System News

Filtered by: Unscheduled Outages | System Notices | (Remove Filters)

• Blue Waters System Reboot in Progress

  Created by: jenos 2021-12-20 10:57:57 (Channels: unscheduledoutages)

   

  The Blue Waters compute system is being rebooted this morning due to an issue with its high speed network that was causing most jobs to stall or fail.  Login node, filesystem, and data mover components are not expected to be interrupted by this operation.

  Thank you for your patience while the issue is resolved.  Estimated return to service for the compute portion is 5pm CT today.

  Watch the Blue Waters portal blog for updates as well as a mail notification upon completion.

  Questions?  Mail help+bw@ncsa.illinois.edu.

  
  Read Less

    The Blue Waters compute system is being rebooted this morning due to an issue with its high speed network that was causing most jobs to stall or fail.  Login node, filesystem, and data mover components are not expected to be interrupted by this operation. Thank you for your patience while the issue...
  Read More

• Blue Waters network maintenance notice

  Created by: jenos 2021-12-15 11:53:52 (Channels: systemnotices)

  Blue Waters partners-

  A network component connecting Blue Waters services will have a brief maintenance operation performed today between 1-2pm CT.  No major service impact is expected and we will be monitoring for connection interrupts, but if you experience connection issues, please report them to help+bw@ncsa.illinois.edu so we can ensure they are resolved. 

  Also, a friendly reminder of the previous notification that Blue Waters service ends on 12/31/2021 and all data transfers need to be completed before that time. Support availability will be limited after 12/23/2021 due to the holiday as well. Please make data transfer plans accordingly.

  Thank you-

  Blue Waters Admins

   

  
  Read Less

  Blue Waters partners- A network component connecting Blue Waters services will have a brief maintenance operation performed today between 1-2pm CT.  No major service impact is expected and we will be monitoring for connection interrupts, but if you experience connection issues, please report them to help+bw@ncsa.illinois.edu so we can ensure...
  Read More

• Blue Waters returned to service

  Created by: jenos 2021-09-28 09:05:33 (Channels: unscheduledoutages)

  Blue Waters has been returned to service as of 9:00AM CT. Running jobs may have experienced I/O suspension if dependent upon the part of the scratch filesystem that was unavailable between 7AM-9AM this morning, or potentially no interruption at all if not. The job scheduler has been resumed.

  Please report any issues to help+bw@ncsa.illinois.edu 

  Blue Waters Admins

  
  Read Less

  Blue Waters has been returned to service as of 9:00AM CT. Running jobs may have experienced I/O suspension if dependent upon the part of the scratch filesystem that was unavailable between 7AM-9AM this morning, or potentially no interruption at all if not. The job scheduler has been resumed. Please report any...
  Read More

• Blue Waters System Outage

  Created by: kingda 2021-09-28 07:58:43 (Channels: unscheduledoutages)

  The Blue Waters Scratch filesystem is partially unavailable due to a power issue.  Login nodes are still accessible, but the scheduler is paused.  Please check the Blue Waters system blog for updates as recovery progresses.

  Thank you for your patience as this is resolved.

  Blue Waters Admins

• Blue Waters Notice: Login Maintenance Update

  Created by: tbouvet 2021-09-25 17:36:50 (Channels: scheduledoutages|systemnotices)

  Update: The login nodes h2ologin1 and h2ologin3 were returned to service September 24th, 4PM CT.

  ---

  The login nodes h2ologin1 and h2ologin3 will be removed from service September 22, 2021 for maintenance. They will have access restricted and current login sessions will drain for 24hrs. Please access the available login nodes by using the round-robin alias bw.ncsa.illinois.edu or h2ologin.ncsa.illinois.edu. A followup email will be sent when they are returned to service.

  Blue Waters Admin Team

  
  Read Less

  Update: The login nodes h2ologin1 and h2ologin3 were returned to service September 24th, 4PM CT. The login nodes h2ologin1 and h2ologin3 will be removed from service September 22, 2021 for maintenance. They will have access restricted and current login sessions will drain for 24hrs. Please access the available login nodes...
  Read More

• Blue Waters Notice : Login Maintenance

  Created by: tbouvet 2021-09-25 09:15:05 (Channels: scheduledoutages|systemnotices)

  The login nodes h2ologin1 and h2ologin3 will be removed from service September 22, 2021 for maintenance. They will have access restricted and current login sessions will drain for 24hrs. Please access the available login nodes by using the round-robin alias bw.ncsa.illinois.edu or h2ologin.ncsa.illinois.edu. A followup email will be sent when they are returned to service.

  Blue Waters Admin Team

  
  Read Less

  The login nodes h2ologin1 and h2ologin3 will be removed from service September 22, 2021 for maintenance. They will have access restricted and current login sessions will drain for 24hrs. Please access the available login nodes by using the round-robin alias bw.ncsa.illinois.edu or h2ologin.ncsa.illinois.edu. A followup email will be sent when...
  Read More

• Blue Waters returned to service

  Created by: jenos 2021-09-21 11:55:53 (Channels: unscheduledoutages)

  Blue Waters has been returned to service as of 11:15AM CT. Most running jobs survived the event, but may have experienced extreme communication speed reduction or even failure in some cases during the event. Existing login sessions were not interrupted, but new login session attempts may have seen intermittent failure during the event.

  Please report any issues to help+bw@ncsa.illinois.edu 

  Blue Waters Admins

  
  Read Less

  Blue Waters has been returned to service as of 11:15AM CT. Most running jobs survived the event, but may have experienced extreme communication speed reduction or even failure in some cases during the event. Existing login sessions were not interrupted, but new login session attempts may have seen intermittent failure...
  Read More

• Blue Waters System Outage

  Created by: jenos 2021-09-21 10:25:10 (Channels: unscheduledoutages)

  The Blue Waters compute resource is currently unavailable due to a power issue.  Login nodes and filesystems are still accessible.  Please check the Blue Waters system blog for updates as recovery progresses.

  Thank you for your patience as this is resolved.

  Blue Waters Admins

• Blue Waters scheduler and login issues

  Created by: kingda 2021-09-21 08:07:18 (Channels: unscheduledoutages)

  The Blue Waters Scheduler and h2ologin3 is currently offline due to a power issue. While this issue persists, jobs cannot be submitted and will not start. Scheduling commands will not be available. No jobs should be lost. Administrators are investigating and will post updates to the system blog.

  Blue Waters Admins

   

• Blue Waters returned to service

  Created by: jenos 2021-08-25 18:47:14 (Channels: unscheduledoutages|systemnotices)

  Blue Waters has been returned to service as of 6:45 pm CT after a compute subsystem reboot to correct a problem with the high speed network. 

  In addition, h2ologin1 has been returned to the service rotation addressed by the [bw|h2ologin].ncsa.illinois.edu hostnames. 

  We appreciate your patience while we addressed these matters.

  The Blue Waters Team

• Blue Waters System Reboot in Progress

  Created by: jenos 2021-08-25 15:46:14 (Channels: unscheduledoutages)

  Update:

  Return to service is now estimated at 6pm CT.

  ---

  The Blue Waters compute system is being rebooted this morning due to an issue with its high speed network that was causing most jobs to fail.  Login node, filesystem, and data mover components are not expected to be interrupted by this operation.  The job scheduler will see a brief interruption for system updates that will overlap with the reboot.

  Thank you for your patience while the issue is resolved.  Estimated return to service for the compute portion is 2pm CT today.

  Watch the Blue Waters portal blog for updates as well as a mail notification upon completion.

  Questions?  Mail help+bw@ncsa.illinois.edu.

  
  Read Less

  Update: Return to service is now estimated at 6pm CT. The Blue Waters compute system is being rebooted this morning due to an issue with its high speed network that was causing most jobs to fail.  Login node, filesystem, and data mover components are not expected to be interrupted by this operation. ...
  Read More

• Blue Waters Notice : Login Maintenance Update

  Created by: tbouvet 2021-08-13 14:35:22 (Channels: scheduledoutages|systemnotices)

  The login nodes h2ologin3 and h2ologin4 have been returned to service.

  The login node h2ologin1 will be removed from production Aug 22, 2021 for general maintenance. It will have access restricted and current login sessions will drain for 24hrs. Please access the available login nodes by using the round-robin alias bw.ncsa.illinois.,edu or h2ologin.ncsa.illinois.edu.

  Blue Waters Admin Team

  
  Read Less

  The login nodes h2ologin3 and h2ologin4 have been returned to service. The login node h2ologin1 will be removed from production Aug 22, 2021 for general maintenance. It will have access restricted and current login sessions will drain for 24hrs. Please access the available login nodes by using the round-robin alias bw.ncsa.illinois.,edu...
  Read More

• Blue Waters Notice : Login Maintenance

  Created by: tbouvet 2021-07-29 16:46:08 (Channels: scheduledoutages|systemnotices)

  The login nodes h2ologin3 and h2ologin4 will be removed from production Aug 9, 2021 for general maintenance. They will have access restricted and current login sessions will drain for 24hrs. Please access the available login nodes by using the round-robin alias bw.ncsa.illinois.,edu or h2ologin.ncsa.illinois.edu. A followup email will be sent when they are returned to service.

  Blue Water Admin Team

  
  Read Less

  The login nodes h2ologin3 and h2ologin4 will be removed from production Aug 9, 2021 for general maintenance. They will have access restricted and current login sessions will drain for 24hrs. Please access the available login nodes by using the round-robin alias bw.ncsa.illinois.,edu or h2ologin.ncsa.illinois.edu. A followup email will be sent...
  Read More

• Blue Waters filesystem issue resolved

  Created by: jenos 2021-03-10 01:36:08 (Channels: unscheduledoutages)

  The Blue Waters scratch filesystem has been serviced and resumed responsiveness by 01:30 AM on 3/10/2021. System impact from this issue may have been observed as early as 05:00 PM on 3/09/2021.  Thank you for being patient while this issue was resolved.

  Blue Waters Admins

   

• Blue Waters filesystem issue

  Created by: jenos 2021-03-10 00:02:29 (Channels: unscheduledoutages)

  The Blue Waters /scratch filesystem is currently encountering an issue causing it to be slow or unresponsive.  While this issue persists, login sessions may hang, and jobs may stop making progress or fail. Administrators are investigating and will post updates to the system blog.

  Blue Waters Admins

   

• Blue Waters Notice: shifter service issue

  Created by: gbauer 2021-02-12 09:06:33 (Channels: unscheduledoutages)

  The issue has been resolved.

  -----

  The shifter container service on Blue Waters is currently experiencing issues when pulling some new images through the shifter gateway.

  Jobs that use currently pulled images should run as expected

  We are looking into the issue and will update the blog periodically., 

  We apologize for the inconvenience.

  -Blue Waters

• H2ologin4 had a system issue that required a reboot to resolve.

  Created by: tbouvet 2020-12-16 15:23:51 (Channels: unscheduledoutages|systemnotices)

  H2ologin4 was rebooted to resolve a system issue that it was experiencing today. We apologize for any inconvenience this may have caused.

• Intermittent Login Issues

  Created by: kingda 2020-12-09 15:54:08 (Channels: unscheduledoutages)

  The Blue Waters login nodes were having issues that were identified and resolved. The issues started at 12:15pm CST and ended at 3:30pm CST. 

  Questions?  Mail help+bw@ncsa.illinois.edu.

• Blue Waters Filesystem Issue Resolved

  Created by: mshow 2020-12-02 20:01:29 (Channels: unscheduledoutages)

  The Blue Waters file system slowness issue has been resolved without a system interruption. The period where jobs or login nodes may have experienced slow filesystem symptoms spanned from 2:44 PM CST 12/2 to 7:55 PM CST 12/2.

  Questions?  Mail help+bw@ncsa.illinois.edu.

• Blue Waters Filesystem Issue

  Created by: mshow 2020-12-02 17:18:56 (Channels: unscheduledoutages)

  The Blue Waters file system is experiencing an issue causing extreme slowness. Admins are investigating the issue and will update the portal blog with any updates prior to resolution. There is no estimated time for return to service yet.

  Questions?  Mail help+bw@ncsa.illinois.edu.

• Blue Waters Filesystem Issue Resolved

  Created by: jenos 2020-11-25 10:05:56 (Channels: unscheduledoutages)

  The Blue Waters file system slowness issue has been resolved without a system interruption. The period where jobs or login nodes may have experienced slow filesystem symptoms spanned from 9:49 PM CST 11/24 to 6:00 AM CST 11/25.

  Questions?  Mail help+bw@ncsa.illinois.edu.

• Blue Waters Filesystem Issue

  Created by: jenos 2020-11-25 02:01:49 (Channels: unscheduledoutages)

  The Blue Waters file system is experiencing an issue causing extreme slowness. Admins are investigating the issue and will update the portal blog with any updates prior to resolution. There is no estimated time for return to service yet.

  Questions?  Mail help+bw@ncsa.illinois.edu.

• Blue Waters returned to service

  Created by: jenos 2020-11-24 17:37:19 (Channels: unscheduledoutages|systemnotices)

  Blue Waters has been returned to service as of 5:35 pm CT after a full system reboot to correct the problem with system boot infrastructure. All previously running jobs were lost.

  The Blue Waters Team

• Blue Waters System Reboot in Progress

  Created by: bbode 2020-11-24 11:14:49 (Channels: unscheduledoutages)

  The Blue Waters compute system is being rebooted this morning due to an issue with its primary boot infrastructure.  Login node, scheduler, filesystem, and data mover components are not expected to be interrupted by this operation.  

  Thank you for your patience while the issue is resolved.  Estimated return to service for the compute portion is 6pm CT today.

  Watch the Blue Waters portal blog for updates as well as a mail notification upon completion.

  Questions?  Mail help+bw@ncsa.illinois.edu.

  
  Read Less

  The Blue Waters compute system is being rebooted this morning due to an issue with its primary boot infrastructure.  Login node, scheduler, filesystem, and data mover components are not expected to be interrupted by this operation.   Thank you for your patience while the issue is resolved.  Estimated return to service for...
  Read More

• Blue Waters upcoming login interrupts

  Created by: tbouvet 2020-11-04 14:13:35 (Channels: scheduledoutages|systemnotices)

  Login Maintenance Schedule

  As logins are rotated out of production, a wall message will be issued as a reminder that an interrupt is pending. h2ologin1 has already been serviced.

  • h2ologin2 access restricted 11/02/2020 10AM
    • Reboot and return to service 11/03/2020 10AM
  • h2ologin3 access restricted 11/02/2020 10AM
    • Reboot and return to service 11/03/2020 10AM
  • h2ologin4 access restricted 11/03/2020 10AM
    • Reboot and return to service 11/04/2020 10AM.

  Note: Aliases bw.ncsa.illinois.edu or h2ologin.ncsa.illinois.edu should reflect the in-service login nodes at any given time.

  UPDATE:11/04/2020 Maintenance is complete on the login nodes.

  
  Read Less

  Login Maintenance Schedule As logins are rotated out of production, a wall message will be issued as a reminder that an interrupt is pending. h2ologin1 has already been serviced. h2ologin2 access restricted 11/02/2020 10AM Reboot and return to service 11/03/2020 10AM h2ologin3 access restricted 11/02/2020 10AM Reboot and return to service 11/03/2020 10AM h2ologin4 access restricted...
  Read More

• Blue Waters returned to service

  Created by: mshow 2020-10-15 18:35:37 (Channels: unscheduledoutages)

  Blue Waters has been returned to service as of 6:30pm CT after a full system reboot to correct the problem on the high-speed network. All previously running jobs were lost.

  The Blue Waters Team

• Blue Waters System Reboot in Progress

  Created by: bbode 2020-10-15 09:49:04 (Channels: unscheduledoutages)

  The Blue Waters compute system is being rebooted this morning due to an HSN issue.  Login node, scheduler, filesystem, and data mover components are not expected to be interrupted by this operation.  During the reboot a set of accumulated security related software updates will be applied. Those updates are not expected to impact users. The Shifter update planned for the 21st will be applied during this downtime as well.

  Thank you for your patience while the issue is resolved.  Estimated return to service for the compute portion is 4pm CT today.

  Watch the Blue Waters portal blog for updates as well as a mail notification upon completion.

  Questions?  Mail help+bw@ncsa.illinois.edu.

  
  Read Less

  The Blue Waters compute system is being rebooted this morning due to an HSN issue.  Login node, scheduler, filesystem, and data mover components are not expected to be interrupted by this operation.  During the reboot a set of accumulated security related software updates will be applied. Those updates are not...
  Read More

• Blue Waters Notice: Software and Maintenance

  Created by: jenos 2020-06-11 12:12:35 (Channels: scheduledoutages|systemnotices|softwareupdate)

  Blue Waters Partners:

  Summary:

  1. Globus command line interface issue is resolved.
  2. Shifter software upgrade on 6/18/2020
  3. Reminder: Portal dynamic data interruption on 6/12/2020

  ---

  1. The Globus issue experienced when using the command line interface provided by bwpy/2.0.2 has been resolved without a need for any local updates, as changes performed by Globus outside of Blue Waters resolved the issue.

  2. On Thursday, June 18, 2020, we will be updating Shifter module from version 16.08.3 to 18.03.4. The new version will provide a number of stability improvements and enables local user images.  When complete, the "shifter" GRES specification will remain valid, but shifter16 will be disabled.

     Documentation will be updated at that point to reflect the changes.

     https://bluewaters.ncsa.illinois.edu/shifter

     Shifter jobs that are already running at that time are unlikely to encounter issues, but it is possible in some circumstances.  If this happens, you can resubmit the job using the "shifter" GRES as documentation will reflect once the change is performed.

  3. A reminder that on Friday June 12 at 2 PM CT, maintenance will be performed on a backend database supporting dynamic system information presented in the Blue Waters portal.  The hardware maintenance should complete in less than an hour. During this timeframe, updated dynamic data regarding Blue Waters system status may not be available. The Blue Waters system is expected to continue to operate normally during this maintenance.

  Please notify help+bw@ncsa.illinois.edu with any questions.

  -Blue Waters project

   

  
  Read Less

  Blue Waters Partners: Summary: Globus command line interface issue is resolved. Shifter software upgrade on 6/18/2020 Reminder: Portal dynamic data interruption on 6/12/2020 The Globus issue experienced when using the command line interface provided by bwpy/2.0.2 has been resolved without a need for any local updates, as changes performed by...
  Read More

• Blue Waters Maintenance Notice

  Created by: mshow 2020-06-09 15:08:42 (Channels: scheduledoutages|systemnotices)

  Blue Waters users-

  On Friday June 12 startnig at 2 PM central, maintenance will be performed on a backend database supporting dynamic system information presented in the Blue Waters portal.  The hardware maintenance should complete in less than an hour. During this timeframe, updated dynamic data regarding Blue Waters system status may not be available. The Blue Waters system is expected to continue to operate normally during this maintenance.

  Questions may be sent to help+bw@ncsa.illinois.edu.

  Thank you for your patience.

  Blue Waters Admins

  
  Read Less

  Blue Waters users- On Friday June 12 startnig at 2 PM central, maintenance will be performed on a backend database supporting dynamic system information presented in the Blue Waters portal.  The hardware maintenance should complete in less than an hour. During this timeframe, updated dynamic data regarding Blue Waters system status may...
  Read More

• Blue Waters returned to service

  Created by: jenos 2020-05-22 14:07:30 (Channels: unscheduledoutages)

  Blue Waters has been returned to service as of 2:00pm CT after a full system reboot to correct the problem on the high-speed network. All previously running jobs were lost.

  The Blue Waters Team

• Blue Waters System Reboot in Progress

  Created by: jenos 2020-05-22 09:05:09 (Channels: unscheduledoutages)

  The Blue Waters compute system is being rebooted this morning due to an unexpected complication with hardware and cooling maintenance.  Login node, scheduler, filesystem, and data mover components are not expected to be interrupted by this operation.  Thank you for your patience while the issue is resolved.  Estimated return to service for the compute portion is 1pm CT today.

  Watch the Blue Waters portal blog for updates as well as a mail notification upon completion.

  Questions?  Mail help+bw@ncsa.illinois.edu.

  
  Read Less

  The Blue Waters compute system is being rebooted this morning due to an unexpected complication with hardware and cooling maintenance.  Login node, scheduler, filesystem, and data mover components are not expected to be interrupted by this operation.  Thank you for your patience while the issue is resolved.  Estimated return to...
  Read More

• Blue Waters Update: RSA deactivation

  Created by: jenos 2020-04-07 16:22:40 (Channels: systemnotices|policychange)

  Blue Waters partners-

  RSA authentication has been deactivated as scheduled, with one exception noted below due to Globus staff availability and illness circumstance. Please note that the Globus Organizational login page will continue to accept RSA, but you may also use your Globus account, as described here.  An updated schedule for it's Duo conversion will be posted when it is known. 

  Relevant Event Timeline:

  December, 2019 

  • Duo authentication enabled
  • "-duo" suffix on supplemental ssh login hostnames and Globus endpoints

  April 1, 2020

  • HPSS Near-line storage system permanently shut down

  April 7, 2020

  • RSA authentication permanently deactivated
  • Original ssh login hostnames and online Globus endpoints will use authentication with Duo instead of RSA
  • "-duo" suffix on supplemental ssh login hostnames deprecated, but will continue to function

  [DELAYED]

  • Globus Online organizational login page will switch from RSA to CILogon
    CILogon Duo authentication can be tested before April 7 at https://cilogon.org/  (use "National Center for Supercomputing Applications" as Identity Provider)

  April 14, 2020

  • Supplemental "-duo" suffixed Globus endpoint names will be permanently retired

  Obsolete RSA hardware tokens do not need to be returned to NCSA, and can be disposed.

  If you have any questions, please send mail to help+bw@ncsa.illinois.edu

   

  
  Read Less

  Blue Waters partners- RSA authentication has been deactivated as scheduled, with one exception noted below due to Globus staff availability and illness circumstance. Please note that the Globus Organizational login page will continue to accept RSA, but you may also use your Globus account, as described here.  An updated schedule for...
  Read More

• Blue Waters Notice: RSA deactivation on 4/7/2020

  Created by: jenos 2020-03-31 10:02:31 (Channels: systemnotices|policychange)

  Blue Waters partners-

   

  As noted previously, RSA authentication to Blue Waters has an upcoming deactivation scheduled.  The replacement authentication method via Duo was activated on Blue Waters in December, 2019. If you have not set up Duo and tested Duo authentication yet (for ssh or Globus endpoints), please do so as it will shift from optional to required for Blue Waters access on April 7th, 2020. The configuration changes required for all of the following events are not expected to cause any service interruption, with the exception of the services that are being retired of course.

  Relevant Event Timeline:

  December, 2019

  • Duo authentication enabled
  • "-duo" suffix on supplemental ssh login hostnames and Globus endpoints

  April 1, 2020

  • HPSS Near-line storage system permanently shut down

  April 7, 2020

  • RSA authentication permanently deactivated
  • Original ssh login hostnames and online Globus endpoints will use authentication with Duo instead of RSA
  • "-duo" suffix on supplemental ssh login hostnames deprecated, but will continue to function
  • Globus Online organizational login page will switch from RSA to CILogon
    CILogon Duo authentication can be tested before April 7 at https://cilogon.org/  (use "National Center for Supercomputing Applications" as Identity Provider)

  April 14, 2020

  • Supplemental "-duo" suffixed Globus endpoint names will be permanently retired

   

  Obsolete RSA hardware tokens do not need to be returned to NCSA, and can be disposed.

  If you have any questions, please send mail to help+bw@ncsa.illinois.edu

   

  
  Read Less

  Blue Waters partners-   As noted previously, RSA authentication to Blue Waters has an upcoming deactivation scheduled.  The replacement authentication method via Duo was activated on Blue Waters in December, 2019. If you have not set up Duo and tested Duo authentication yet (for ssh or Globus endpoints), please do so as...
  Read More

• Blue Waters returned to service

  Created by: kingda 2020-03-29 18:41:43 (Channels: unscheduledoutages|systemnotices)

  Blue Waters has been returned to service after a full system reboot to correct a problem on the high-speed network. All previously running jobs were lost.

  The Blue Waters Team

• Blue Waters Unscheduled Outage

  Created by: jenos 2020-03-29 13:01:11 (Channels: unscheduledoutages)

  Blue Waters is experiencing an issue with the high-speed network that began at 12:02 PM CT. System support staff are evaluating and attempting to restore normal service. Job scheduling is paused until the issue is resolved. The file systems and data transfer services are operating normally. Interim updates will be posted on the Blue Waters Message of the Day: https://bluewaters.ncsa.illinois.edu/motd

  Current return to service estimate (optimal): 7:00pm CT

  We appreciate your patience while this is resolved.

   

  
  Read Less

  Blue Waters is experiencing an issue with the high-speed network that began at 12:02 PM CT. System support staff are evaluating and attempting to restore normal service. Job scheduling is paused until the issue is resolved. The file systems and data transfer services are operating normally. Interim updates will be posted on...
  Read More

• Blue Waters returned to service

  Created by: bbode 2020-03-16 16:07:00 (Channels: unscheduledoutages|systemnotices)

  Blue Waters has been returned to service after a full system reboot to correct a problem on the high-speed network. All previously running jobs were lost.

  The Blue Waters Team

• Blue Waters Notice: Programming Environments changed

  Created by: gbauer 2020-01-23 17:59:03 (Channels: systemnotices)

  Blue Waters partners;

  As announced back in December, the version defaults for the 4 programming environments (cray, gnu, intel and pgi) have been changed to newer versions as of 1/23/2020. A table with the changes is available at https://bluewaters.ncsa.illinois.edu/pe-updates#pe18.12. Note that gcc moved from 4.9.3 to 6.3.0 which can impact C++ codes and libraries.

  For users who wish to continue with the older GNU compiler please use PrgEnv-gnu/5.2.82-gcc.4.9.3 as in:

  $ module unload PrgEnv-cray/5.2.82
$ module load PrgEnv-gnu/5.2.82-gcc.4.9.3

  and then select the older modules listed in the table at https://bluewaters.ncsa.illinois.edu/pe-updates#pe18.12.

  Statically linked applications should continue to run without needing to be recompiled. Execution of dynamically linked applications should be checked. You can use the ldd command to check the linkage model your executable. Monitor your jobs already in the queue.

  If you encounter issues while testing the new environments please send email to help+bw@ncsa.illinois.edu.

  ----

  To prepare for the change we have provided metamodules to make your environment close to what the defaults will be.To test the new programming environments please follow the instructions provided at https://bluewaters.ncsa.illinois.edu/pe-updates#pe18.12. Some known issues are documented as well.

   

  
  Read Less

  Blue Waters partners; As announced back in December, the version defaults for the 4 programming environments (cray, gnu, intel and pgi) have been changed to newer versions as of 1/23/2020. A table with the changes is available at https://bluewaters.ncsa.illinois.edu/pe-updates#pe18.12. Note that gcc moved from 4.9.3 to 6.3.0 which can impact C++ codes...
  Read More

• Blue Waters Reminder: Programming Environment change testing period 12/23/2019 to 01/23/2020

  Created by: gbauer 2020-01-22 20:50:24 (Channels: systemnotices)

  Blue Waters partners;

  This is a reminder that on January 23rd  2020 the version defaults for the 4 programming environments (cray, gnu, intel and pgi) will change to newer versions. Statically linked applications should continue to run without needing to be recompiled. Execution of dynamically linked applications should be checked. You can use the ld command to check the linkage model your executable. Monitor your jobs already in the queue.

  If you encounter issues while testing the new environments please send email to help+bw@ncsa.illinois.edu.

  ----

  To prepare for the change we have provided metamodules to make your environment close to what the defaults will be.To test the new programming environments please follow the instructions provided at https://bluewaters.ncsa.illinois.edu/pe-updates#pe18.12. Some known issues are documented as well.

   

  
  Read Less

  Blue Waters partners; This is a reminder that on January 23rd  2020 the version defaults for the 4 programming environments (cray, gnu, intel and pgi) will change to newer versions. Statically linked applications should continue to run without needing to be recompiled. Execution of dynamically linked applications should be checked. You can use...
  Read More

• Reminder: Blue Waters System Changes

  Created by: jenos 2020-01-14 00:08:02 (Channels: systemnotices|policychange|newfunctionality)

  Blue Waters Users:

  In December 2019, you received a notification detailing some planned system changes.

  Transition summary: https://bluewaters.ncsa.illinois.edu/transition

  January 14, 2020 is the planned date to effect two changes worth noting:

  1. SSH host key change for those using RSA-based access to login nodes
     Accessing logins via ssh for some may yield an alarming warning which includes text like:
     @@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@
@    WARNING: REMOTE HOST IDENTIFICATION HAS CHANGED!     @
@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@
IT IS POSSIBLE THAT SOMEONE IS DOING SOMETHING NASTY!
Someone could be eavesdropping on you right now (man-in-the-middle attack)!
     This is a normal warning to receive when a host key is changed.  Please follow the instructions in the FAQ to resolve the warning. Current host key fingerprints are posted there for verification.

  2. Filesystem ACL Refresh
     Home, Projects, and Scratch directories will no longer allow permission (ACL) broadening by default. A service request may be created for admins to configure a specific ACL or enable self-management. This change may be disruptive to workflows depending on access to another user's folders unless a service request is made.  Detailed information on the change and how to request exceptions is available here.

  Blue Waters Admins

   

  
  Read Less

  Blue Waters Users: In December 2019, you received a notification detailing some planned system changes. Transition summary: https://bluewaters.ncsa.illinois.edu/transition January 14, 2020 is the planned date to effect two changes worth noting: SSH host key change for those using RSA-based access to login nodes Accessing logins via ssh for some may yield an alarming warning...
  Read More

• BlueWaters returned to service

  Created by: mshow 2019-12-22 23:32:00 (Channels: mandatory|unscheduledoutages)

  After a full system reboot and checkout, the system hs been retrned to full service operations

• BlueWaters system reboot

  Created by: mshow 2019-12-22 13:55:58 (Channels: unscheduledoutages|systemnotices)

  While the filesystem issue has been resolved, a full system reboot will be required before returning to production status. It is our expectation that the system will return later this evening.

  BW Admin Team

• Blue Waters: UPDATE Scheduler Remains Paused

  Created by: tbouvet 2019-12-22 08:21:11 (Channels: unscheduledoutages|systemnotices)

  UPDATE: Scheduler Remains Paused as we continue to restore the scratch file system to service.

  The Blue Waters scheduler is currently paused due to a meda data server issue with the scratch file system. We are actively working the issue and new logins will likely hang without completion.

  Status updates will be posted to the blog on the Blue Waters portal.

  https://bluewaters.ncsa.illinois.edu/message-of-the-day

  The Blue Waters team.

  
  Read Less

  UPDATE: Scheduler Remains Paused as we continue to restore the scratch file system to service. The Blue Waters scheduler is currently paused due to a meda data server issue with the scratch file system. We are actively working the issue and new logins will likely hang without completion. Status updates will be...
  Read More

• Blue Waters: Scheduler paused because of a scratch file system issue.

  Created by: tbouvet 2019-12-21 20:06:21 (Channels: unscheduledoutages|systemnotices)

  The Blue Waters scheduler is currently paused due to a meda data server issue with the scratch file system. We are actively working the issue and new logins will likely hang without completion.

  Status updates will be posted to the blog on the Blue Waters portal.

  The Blue Waters team.

• Blue Waters Returned to Service

  Created by: tbouvet 2019-12-21 19:13:07 (Channels: unscheduledoutages|systemnotices)

  Blue Waters returned to service 12/21/2019 at 7:00PM following today's file system issue.

  Please email help+bw@ncsa.illinois.edu to report any issues.

   

• Blue Waters: Scheduler paused due to file system issue

  Created by: bbode 2019-12-21 13:08:13 (Channels: unscheduledoutages|systemnotices)

  The Blue Waters scheduler is currently paused due to two down storage targets in the scratch file system. Staff are currently working to resolve the issue. 

  Status updates will be posted to the blog on the Blue Waters portal.

  The Blue Waters team.

• Blue Waters Operations Transitioning

  Created by: jenos 2019-12-19 19:03:08 (Channels: mandatory|systemnotices|general|policychange)

  Hello Blue Waters partners,

  Today marks the conclusion of regular NSF Blue Waters operations and allocations.  Allocations ending today will be granted the normal 90 day grace period to transfer data off Blue Waters storage systems. After today, jobs remaining in the queue will be permitted to run but job submission will be disabled for expired projects.

  As Blue Waters transitions to support changing workflow needs for the NGA phase of operations, several policies and processes will be adjusted. Though they are detailed in the summary page linked below, additional notices will be issued near key timeframes.

  Transition summary: https://bluewaters.ncsa.illinois.edu/transition

  If you have questions about any changes, please submit a service request as normal to help+bw@ncsa.illinois.edu, but keep in mind that the upcoming holiday period may not have typical business hour response.

  As a final note, the Blue Waters staff has thoroughly enjoyed the challenges, innovations, enthusiasm, and breakthrough science experienced over the last seven years with everyone involved. Best of luck in the next stage of your research!

  
  Read Less

  Hello Blue Waters partners, Today marks the conclusion of regular NSF Blue Waters operations and allocations.  Allocations ending today will be granted the normal 90 day grace period to transfer data off Blue Waters storage systems. After today, jobs remaining in the queue will be permitted to run but job submission...
  Read More

• Blue Waters Returned to Service

  Created by: bbode 2019-11-28 22:36:06 (Channels: unscheduledoutages|systemnotices)

  Blue Waters has been rebooted and returned to service at 10:35PM following an issue with the high-speed network earlier this afternoon. All running jobs were lost due to the outage.

  Please email help+bw@ncsa.illinois.edu to report any issues.

• Blue Waters: HSN issues full reboot in progress

  Created by: bbode 2019-11-28 17:59:37 (Channels: unscheduledoutages|systemnotices)

  An issue with the high-speed network on Blue Waters has forced a full system reboot. We currently anticipate a return to service of 11PM CST. 

  Status updates will be posted to the blog on the Blue Waters portal.

  The Blue Waters team.

• Nearline Tape Library Has Returned To Service

  Created by: briandi 2019-11-14 17:19:54 (Channels: unscheduledoutages|systemnotices)

  The NCSA_Nearline storage subsystem issue on Blue Waters was resolved and the system returned to normal operations at 3:30 pm.

   

• Nearline Tape Library Emergency System Maintenance

  Created by: briandi 2019-11-14 12:06:32 (Channels: unscheduledoutages|systemnotices)

  Blue Waters Users,

  Blue Waters is experiencing an issue on a subset of the HPSS storage subsystem (ncsa#Nearline) that began early this morning. System support staff are evaluating and attempting to restore it to normal service. The rest of Blue Waters subsystems remain in normal operation.

  Some transfers in and out of the system may be affected.

  Interim updates will be posted on the Blue Waters Message of the Day: https://bluewaters.ncsa.illinois.edu/motd

   

  
  Read Less

  Blue Waters Users, Blue Waters is experiencing an issue on a subset of the HPSS storage subsystem (ncsa#Nearline) that began early this morning. System support staff are evaluating and attempting to restore it to normal service. The rest of Blue Waters subsystems remain in normal operation. Some transfers in and out of...
  Read More

• Blue Waters notice: disabling scheduler support for minimum wall clock (minwclimit) jobs

  Created by: gbauer 2019-10-09 17:37:28 (Channels: systemnotices)

  Blue Waters partners;

  We have recently discovered an issue in the batch job scheduler's support for minimum wall clock (minwclimit) jobs. Such jobs express wall clock flexibility with both a minimum wall clock and a preferred wall clock to improve job throughput. The issue is not expected to be resolved in a timely manner. 

  Effective now, already queued jobs with the minwclimit specifier and future submitted jobs with the minwclimit specifier will only run to the wall clock time specified by minwclimit. 

  After October 16th, jobs submitted with the minwclimit specifier will be rejected by the submit filter with the appropriate warning that minwclimit is no longer supported. An announcement will be made if and when support for minwclimit can be reenabled.  

  We apologize for the inconvenience.

  Please send comments or questions to help+bw@ncsa.illinois.edu.

  Blue Waters Project

  
  Read Less

  Blue Waters partners; We have recently discovered an issue in the batch job scheduler's support for minimum wall clock (minwclimit) jobs. Such jobs express wall clock flexibility with both a minimum wall clock and a preferred wall clock to improve job throughput. The issue is not expected to be resolved in a timely...
  Read More

• Blue Waters Returned to Service

  Created by: kingda 2019-10-01 20:46:38 (Channels: scheduledoutages|unscheduledoutages|systemnotices)

  Blue Waters returned to service at 8:45PM following today's scheduler maintenance. 

  Please email help+bw@ncsa.illinois.edu to report any issues.

• Scheduled systems testing period extended to 8PM Central

  Created by: jenos 2019-10-01 16:54:57 (Channels: scheduledoutages|unscheduledoutages|systemnotices)

  Update: This testing period will be extended to 8PM due to unanticipated system-related delays.

  ---

  A scheduled systems testing period will take place on Tuesday, October 1st from 7AM to 5PM 8PM Central, necessitating a shutdown of the job scheduler. Compute nodes will not be available during the test period. Login nodes and file systems will remain accessible.

  
  Read Less

  Update: This testing period will be extended to 8PM due to unanticipated system-related delays. A scheduled systems testing period will take place on Tuesday, October 1st from 7AM to 5PM 8PM Central, necessitating a shutdown of the job scheduler. Compute nodes will not be available during the test period. Login nodes and file...
  Read More

• Blue Waters outage

  Created by: mshow 2019-10-01 10:15:54 (Channels: unscheduledoutages|systemnotices)

  Multiple cabinets have failed within the Blue Waters system. The failed area will be bypassed and operations will continue. 

• Reminder: Scheduled systems testing period on Tuesday, October 1st from 7AM to 5PM Central

  Created by: kingda 2019-09-30 16:44:24 (Channels: scheduledoutages|systemnotices)

  A scheduled systems testing period will take place on Tuesday, October 1st from 7AM to 5PM Central, necessitating a shutdown of the job scheduler. Compute nodes will not be available during the test period. Login nodes and file systems will remain accessible.

• Scheduled systems testing period on Tuesday, October 1st from 7AM to 5PM Central

  Created by: kingda 2019-09-23 20:36:26 (Channels: scheduledoutages|systemnotices)

  A scheduled systems testing period will take place on Tuesday, October 1st from 7AM to 5PM Central, necessitating a shutdown of the job scheduler. Compute nodes will not be available during the test period. Login nodes and file systems will remain accessible.

• Blue Waters Returned to Service

  Created by: kingda 2019-09-12 16:28:51 (Channels: scheduledoutages|systemnotices)

  Blue Waters returned to service at 4:00PM following today's scheduler maintenance. 

  Please email help+bw@ncsa.illinois.edu to report any issues.

• Scheduled systems testing period on Thursday, September 12th from 10AM to 5PM Central

  Created by: gbauer 2019-09-12 09:51:55 (Channels: scheduledoutages|systemnotices)

  A scheduled systems testing period will take place on Thursday, September 12th from 10AM to 5PM Central,  necessitating a shutdown of the job scheduler. Compute nodes will not be available during the test period. Login nodes and file systems will remain accessible.

• Blue Waters: NPCF Power Issue Update

  Created by: tbouvet 2019-07-06 03:45:32 (Channels: unscheduledoutages|systemnotices)

  Blue Waters: NPCF Power Issue 7/5/2019 3PM

  All Blue Waters Resources are available except for the compute nodes. Blue Waters Computes are being rebooted and all running jobs were lost. No RTS eta yet.

• Blue Waters Returned to Service

  Created by: bbode 2019-07-05 17:57:32 (Channels: unscheduledoutages|systemnotices)

  Blue Waters has been rebooted and returned to service at 5:55PM following a power interuption earlier this afternoon. All running jobs were lost due to the outage.

  Please email help+bw@ncsa.illinois.edu to report any issues.

• Blue Waters: NPCF Power Issue, Scheduler paused expect full reboot

  Created by: tbouvet 2019-07-05 13:32:15 (Channels: unscheduledoutages|systemnotices)

  A power outage at the building housing the Blue Waters system has caused a service interruption; the Login Nodes, Network, Storage, Compute Nodes, and Near-line Storage may be unavailable. It is unknown at this time when a return to service can be expected. Watch the Blue Waters portal blog for updates.

• Blue Waters: NPCF Facility Power Maintenance Complete

  Created by: tbouvet 2019-06-24 06:07:40 (Channels: scheduledoutages|systemnotices)

  The NPCF facility power maintenance is complete and Blue Waters has returned to service at 06:00 hours.

  Please email help+bw@ncsa.illinois.edu to report any issues.

   

• Blue Waters Scheduled Maintence Extended until June 24 6AM CT

  Created by: tbouvet 2019-06-24 05:18:12 (Channels: scheduledoutages|systemnotices)

  The scheduled maintenance has been extended until June 24th 6AM due to technical difficulties. We apologize for the delay in Return to Service.

  Availability details:

  Subsystem / Services	Projected Interruption Duration
  Compute	June 23 05:00 - June 24 06:00
  Login / Filesystem	June 23 05:00 - June 24 06:00
  Scheduler	June 23 05:00 - June 24 06:00
  Globus online endpoint (ncsa#BlueWaters)	June 23 05:00 - June 24 05:00
  Globus nearline endpoint (ncsa#Nearline)	June 23 05:00 - June 24 05:00

   

  The Blue Waters blog will be updated with any status changes. Send any questions or concerns to help+bw@ncsa.illinois.edu.

  
  Read Less

  The scheduled maintenance has been extended until June 24th 6AM due to technical difficulties. We apologize for the delay in Return to Service. Availability details: Subsystem / Services  Projected Interruption Duration Compute June 23 05:00 - June 24 06:00 Login / Filesystem June 23 05:00 - June 24 06:00 Scheduler June 23 05:00 - June 24...
  Read More

• Reminder: Blue Waters Scheduled Maintenance June 23rd 2019 5AM

  Created by: tbouvet 2019-06-22 14:39:23 (Channels: scheduledoutages|systemnotices)

  Reminder of Maintenance,

  On Sunday June 23, electrical maintenance is scheduled to take place at the NPCF Building which hosts Blue Waters.  Therefore, beginning at 05:00 CT, all Blue Waters subsystems will be unavailable.  The outage is expected to last until 03:00 the following morning, June 24.

  During this interruption, Blue Waters and its subsystems will also receive several accumulated software updates, and minor hardware repairs. It's possible that certain subsystems may return to service ahead of schedule, but the current estimated availability is as follows:

  Availability details:

  Subsystem / Services	Projected Interruption Duration
  Compute	June 23 05:00 - June 24 03:00
  Login / Filesystem	June 23 05:00 - June 24 03:00
  Scheduler	June 23 05:00 - June 24 03:00
  Globus online endpoint (ncsa#BlueWaters)	June 23 05:00 - June 24 03:00
  Globus nearline endpoint (ncsa#Nearline)	June 23 05:00 - June 24 03:00

   

  The Blue Waters blog will be updated with any status changes. Send any questions or concerns to help+bw@ncsa.illinois.edu.

   

  
  Read Less

  Reminder of Maintenance, On Sunday June 23, electrical maintenance is scheduled to take place at the NPCF Building which hosts Blue Waters.  Therefore, beginning at 05:00 CT, all Blue Waters subsystems will be unavailable.  The outage is expected to last until 03:00 the following morning, June 24. During this interruption, Blue Waters and...
  Read More

• Blue Waters: Nearline Tape Library Has Returned To Service

  Created by: tbouvet 2019-05-31 12:58:05 (Channels: unscheduledoutages|systemnotices)

  The Nearline subsystem reboot completed at 11:30AM today.

  All existing transfers should resume as Nearline was returned to service.

• Nearline Tape Library System Reboot

  Created by: bbode 2019-05-31 09:28:04 (Channels: unscheduledoutages|systemnotices)

  The Nearline subsystem is currently undergoing an emergancy reboot to clear multiple issues. It is expected to return to service by 1PM today.

  All existing transfers will resume once Nearline returns to service.

• Blue Waters Has Returned to Service

  Created by: tbouvet 2019-05-16 17:15:55 (Channels: unscheduledoutages|systemnotices)

  The storage issue on Blue Waters projects file system has been resolved and the system returned to normal operations at "5:07" PM CT. Any teams who were impacted by the file system issue have been contacted individually. The scheduler has resumed normal operations.

  Thank you for your patience while this was repaired.

   

  
  Read Less

  The storage issue on Blue Waters projects file system has been resolved and the system returned to normal operations at "5:07" PM CT. Any teams who were impacted by the file system issue have been contacted individually. The scheduler has resumed normal operations. Thank you for your patience while this was...
  Read More

• Blue Waters Project File System Update

  Created by: tbouvet 2019-05-16 09:29:40 (Channels: unscheduledoutages|systemnotices)

  We are in the process of running a file system check and repair of a small portion of the projects file system. When that is complete we will access the results and take appropriate action. Update: File system repair continues and is expected to last until late this afternoon (5/16).

  If you have jobs that do not use the projects file system please email help+bw@ncsa.illinois.edu to let us know and we can release your jobs to run.

  
  Read Less

  We are in the process of running a file system check and repair of a small portion of the projects file system. When that is complete we will access the results and take appropriate action. Update: File system repair continues and is expected to last until late this afternoon (5/16). If...
  Read More

• Blue Waters Project File System Issue

  Created by: tbouvet 2019-05-15 15:20:25 (Channels: unscheduledoutages|systemnotices)

  Blue Waters Users,

  Blue Waters is currently experiencing a storage server issue for a portion of the projects filesystem. As a result, all I/O transactions targeting the affected storage server will block. A single storage server supplies a small fraction of the file system data and all remaining storage servers continue normal operation. The rest of Blue Waters including the other filesystems are operating normally. We have paused the scheduler while we investigate the issue. Interim updates will be posted on the Blue Waters Message of the Day: https://bluewaters.ncsa.illinois.edu/motd
   

  
  Read Less

  Blue Waters Users, Blue Waters is currently experiencing a storage server issue for a portion of the projects filesystem. As a result, all I/O transactions targeting the affected storage server will block. A single storage server supplies a small fraction of the file system data and all remaining storage servers continue normal operation....
  Read More

• Blue Waters OSS Return to Service

  Created by: squaire3 2019-05-12 12:33:07 (Channels: unscheduledoutages|systemnotices)

  The storage issue on Blue Waters was resolved and the system returned to normal operations at 12:30 PM CT. Scheduler has also been resumed.

• Blue Waters OSS Failover

  Created by: squaire3 2019-05-12 09:18:53 (Channels: unscheduledoutages|systemnotices)

  Blue Waters is currently experiencing a storage server failover on the OSS file system that began at 8:08 AM CT. As a result, all I/O transactions targeting the affected storage server will block until the failover completes on all clients. A single storage server supplies a small fraction of the file system data and all remaining storage servers continue normal operation. The rest of Blue Waters including the other filesystems are operating normally. Recovery from this type event (without complications) is expected to take less than two hours. The scheduler will be paused during this time.

  
  Read Less

  Blue Waters is currently experiencing a storage server failover on the OSS file system that began at 8:08 AM CT. As a result, all I/O transactions targeting the affected storage server will block until the failover completes on all clients. A single storage server supplies a small fraction of the...
  Read More

• Nearline Tape Library System Maintenance

  Created by: glasgow 2019-04-29 16:55:33 (Channels: unscheduledoutages|systemnotices)

  Nearline maintenance work is complete and the service has returned to full operations as of:

  1600hrs, April 29th, 2019

  ---

  Nearline is undergoing emergency service on one tape library. The work is related to fallout from last week's hardware service and is expected to take approximately 10 hours to complete. Some files may not be accessible during that time.

  Start time: 0945hrs to ~ 2000hrs, April 29, 2019

  
  Read Less

  Nearline maintenance work is complete and the service has returned to full operations as of: 1600hrs, April 29th, 2019 --- Nearline is undergoing emergency service on one tape library. The work is related to fallout from last week's hardware service and is expected to take approximately 10 hours to complete. Some files may...
  Read More

• Nearline Tape Library System Maintenance

  Created by: glasgow 2019-04-29 09:42:49 (Channels: unscheduledoutages|systemnotices)

  Nearline is undergoing emergency service on one tape library. The work is related to fallout from last week's hardware service and is expected to take approximately 10 hours to complete. Some files may not be accessible during that time.

  Start time: 0945hrs to ~ 2000hrs, April 29, 2019

• Nearline Tape Library System Maintenance

  Created by: glasgow 2019-04-22 15:04:14 (Channels: scheduledoutages|systemnotices)

  One of Nearline's four tape libraries will undergo maintenance to correct power control problems. This library system will be unavailable for four hours while the work is conducted. Files stored in the library will be unavailable for staging during that time. Retrieval jobs from Globus will wait for the files to become available again and continue. No user intervention is necessary.

  Time of outage 0800-1200hrs, Friday, April 26, 2019.

  
  Read Less

  One of Nearline's four tape libraries will undergo maintenance to correct power control problems. This library system will be unavailable for four hours while the work is conducted. Files stored in the library will be unavailable for staging during that time. Retrieval jobs from Globus will wait for the files...
  Read More

• Blue Waters Return to Service

  Created by: squaire3 2019-04-12 15:41:38 (Channels: unscheduledoutages|systemnotices)

  The high-speed network issue on Blue Waters was resolved and the system returned to normal operations at 3:34 PM CT. Scheduler has also been resumed.

• Blue Waters Scheduler Paused - HSN Issue

  Created by: squaire3 2019-04-12 15:02:55 (Channels: unscheduledoutages|systemnotices)

  Blue Waters is experiencing an issue on the high-speed network that began at 2:10 PM CT. System support staff are evaluating and attempting to restore normal service. Job scheduling is paused until the issue is resolved. The file systems and data transfer services are operating normally. Logins have been occationally hanging due to this issue.

  We are expecting the issue to be resolved shortly.

  
  Read Less

  Blue Waters is experiencing an issue on the high-speed network that began at 2:10 PM CT. System support staff are evaluating and attempting to restore normal service. Job scheduling is paused until the issue is resolved. The file systems and data transfer services are operating normally. Logins have been occationally hanging...
  Read More

• Nearline Endpoint Paused for Storage Maintenance

  Created by: glasgow 2019-04-08 23:14:20 (Channels: unscheduledoutages|systemnotices)

  The Nearline endpoint has now been returned to normal operations. 

  The Blue Water's Nearline endpoint will be paused beginning at 1700hrs CDT. New and current user actions/requests will be paused and will resume normal activity when the endpoint is released. No user action is necessary. This maintenance window will be used to conduct resource management operations that have been deferred due to heavy system load. The system will return to normal service no-later-than 2300hrs (5PM-11PM, April 8th, 2019).

  
  Read Less

  The Nearline endpoint has now been returned to normal operations.  The Blue Water's Nearline endpoint will be paused beginning at 1700hrs CDT. New and current user actions/requests will be paused and will resume normal activity when the endpoint is released. No user action is necessary. This maintenance...
  Read More

• Nearline Endpoint Paused for Storage Maintenance

  Created by: glasgow 2019-04-08 16:27:20 (Channels: unscheduledoutages|systemnotices)

  The Blue Water's Nearline endpoint will be paused beginning at 1700hrs CDT. New and current user actions/requests will be paused and will resume normal activity when the endpoint is released. No user action is necessary. This maintenance window will be used to conduct resource management operations that have been deferred due to heavy system load. The system will return to normal service no-later-than 2300hrs (5PM-11PM, April 8th, 2019).

  
  Read Less

  The Blue Water's Nearline endpoint will be paused beginning at 1700hrs CDT. New and current user actions/requests will be paused and will resume normal activity when the endpoint is released. No user action is necessary. This maintenance window will be used to conduct resource management operations that have been deferred...
  Read More

• Blue Waters Returned to Service

  Created by: tbouvet 2019-04-07 15:14:36 (Channels: unscheduledoutages|systemnotices)

  Blue WAters Users,

  The storage server issue on the scratch file system is resolved. I/O transactions initiated during the outage should have resumed when the Lustre target returned. Blue Waters has resumed normal operations.

• Blue Waters Scheduler is paused 9:30 AM

  Created by: tbouvet 2019-04-07 10:30:00 (Channels: unscheduledoutages|systemnotices)

  Blue Waters Users,

  Blue Waters is currently experiencing a storage server issue for a small portion of the scratch filesystem. As a result, all I/O transactions targeting the affected storage server will block. A single storage server supplies a small fraction of the file system data and all remaining storage servers continue normal operation. The rest of Blue Waters including the other filesystems are operating normally. We have paused the scheduler while we investigate the issue. Interim updates will be posted on the Blue Waters Message of the Day: https://bluewaters.ncsa.illinois.edu/motd

  
  Read Less

  Blue Waters Users, Blue Waters is currently experiencing a storage server issue for a small portion of the scratch filesystem. As a result, all I/O transactions targeting the affected storage server will block. A single storage server supplies a small fraction of the file system data and all remaining storage servers continue normal...
  Read More

• BlueWaters cabinet failure

  Created by: mshow 2019-03-23 16:44:36 (Channels: unscheduledoutages|systemnotices)

  The cabinet has been restored.

• BlueWaters repair complete

  Created by: mshow 2019-03-23 05:13:22 (Channels: unscheduledoutages|systemnotices)

  The cabinet has been restored

• BlueWaters cabinet failure

  Created by: mshow 2019-03-23 03:46:37 (Channels: unscheduledoutages|systemnotices)

  A cabinet has shutdown resulting in job loss and an incomplete network configuration. It is unknown at this time when a return to service can be expected for that cabinet. Watch the Blue Waters portal blog for updates.

• Blue Waters returned to service

  Created by: mshow 2019-02-24 03:41:58 (Channels: mandatory|unscheduledoutages|systemnotices)

  The high-speed network issue on Blue Waters was resolved and the system returned to normal operations at 3:30 AM CT.

• Blue Waters High Speed Network issue

  Created by: mshow 2019-02-23 23:43:19 (Channels: unscheduledoutages|systemnotices)

  Blue Waters is experiencing an issue on the high-speed network that began at 9:48 PM CT. System support staff are evaluating and attempting to restore normal service. Job scheduling is paused until the issue is resolved. The file systems and data transfer services are operating normally. Interim updates will be posted on the Blue Waters Message of the Day: https://bluewaters.ncsa.illinois.edu/motd

  
  Read Less

  Blue Waters is experiencing an issue on the high-speed network that began at 9:48 PM CT. System support staff are evaluating and attempting to restore normal service. Job scheduling is paused until the issue is resolved. The file systems and data transfer services are operating normally. Interim updates will be posted on...
  Read More

• Blue Waters Notice: System returned to service

  Created by: jenos 2019-02-06 15:31:27 (Channels: unscheduledoutages)

  Blue Waters Users:

  The reboot is complete and the system has returned to service as of 3:14pm CT. Tomorrow's near-line storage maintenance will proceed as planned.

  We apologize for any inconvenience.

   

• Blue Waters Unplanned Reboot

  Created by: bbode 2019-02-06 10:00:16 (Channels: unscheduledoutages)

  Blue Waters Users,

  We experienced an issue that has the high speed network in an unrecoverable state. We have to reboot the system to recover and all running jobs will be lost. The login nodes and endpoints (ncsa#Nearline ncsa#BlueWaters) will remain available during the reboot. The current estimate for return to service is 5PM CST. Watch the Blue Waters portal blog for updates.

  
  Read Less

  Blue Waters Users, We experienced an issue that has the high speed network in an unrecoverable state. We have to reboot the system to recover and all running jobs will be lost. The login nodes and endpoints (ncsa#Nearline ncsa#BlueWaters) will remain available during the reboot. The current estimate for return to...
  Read More

• Blue Waters Notice: System returned to service

  Created by: tbouvet 2019-01-12 13:09:15 (Channels: scheduledoutages|unscheduledoutages)

  Blue Waters Users:

  The reboot is complete and the system has returned to service.

  We apologize for any inconvenience.

   

• Blue Waters Notice: Unplanned Reboot

  Created by: tbouvet 2019-01-12 09:58:18 (Channels: unscheduledoutages|systemnotices)

  Blue Waters Users,

  We experienced an issue that has the high speed network in an unrecoverable state. We have to reboot the system to recover and all running jobs will be lost. The login nodes and endpoints (ncsa#Nearline ncsa#BlueWaters) will remain available during the reboot. The current estimate for return to service would be this afternoon. Watch the Blue Waters portal blog for updates.

   

  
  Read Less

  Blue Waters Users, We experienced an issue that has the high speed network in an unrecoverable state. We have to reboot the system to recover and all running jobs will be lost. The login nodes and endpoints (ncsa#Nearline ncsa#BlueWaters) will remain available during the reboot. The current estimate for return to...
  Read More

• Blue Waters Notice: Nearline System (HPSS) return to service

  Created by: tbouvet 2019-01-10 15:24:20 (Channels: unscheduledoutages|systemnotices)

  Blue Waters Users:

  The ncsa#Nearline endpoint is available and has returned to service at 3PM CT. We apologize for any inconvenience.

  -Blue Waters

• Blue Waters Notice: Nearline System (HPSS) remains unavailable.

  Created by: tbouvet 2019-01-10 09:38:30 (Channels: unscheduledoutages|systemnotices)

  Blue Waters Users:

  The ncsa#Nearline endpoint is paused pending repairs suffered to systems during the data center power failure on 1/09/19. We apologize for any inconvenience.

  Please check back for an update on the situtation.

  -Blue Waters

   

• Blue Waters Notice: System has returned to service

  Created by: jenos 2019-01-10 09:37:52 (Channels: unscheduledoutages|systemnotices)

   

  Blue Waters Users:

  The power outage at the NPCF building has been resolved.  All Blue Waters services are now available with the exception of the Nearline storage system, which will take a bit longer to recover.  The job scheduler has been resumed and login nodes have access re-enabled.  There will be another notification when the Nearline storage comes back online. We apologize for any inconvenience.

   

  
  Read Less

    Blue Waters Users: The power outage at the NPCF building has been resolved.  All Blue Waters services are now available with the exception of the Nearline storage system, which will take a bit longer to recover.  The job scheduler has been resumed and login nodes have access re-enabled.  There will be...
  Read More

• Blue Waters Notice: Power disruption

  Created by: jenos 2019-01-09 16:21:59 (Channels: unscheduledoutages|systemnotices)

  Blue Waters Users:

  ---

  Update 4:20pm:

  All systems up except Nearline.  Access remains restricted while performance tests complete.

  ---

  A power outage at the building housing the Blue Waters system has caused a service interruption; all running jobs were consequently terminated.  All Blue Waters subsystems have been affected and are currently out of service. The job scheduler is currently paused.  There is no current estimate for return to service, but best case return time would be this evening. Watch the Blue Waters portal blog for updates.

  
  Read Less

  Blue Waters Users: Update 4:20pm: All systems up except Nearline.  Access remains restricted while performance tests complete. A power outage at the building housing the Blue Waters system has caused a service interruption; all running jobs were consequently terminated.  All Blue Waters subsystems have been affected and are currently out of...
  Read More

• Blue Waters Notice: System returned to service

  Created by: jenos 2018-12-27 23:43:11 (Channels: unscheduledoutages|systemnotices)

  Blue Waters Users:

  The power outage at the Blue Waters building has been resolved and the compute nodes have been restarted. The Blue Waters system has been returned to normal operations at 11:40 PM CT. We apologize for any inconvenience.

   

• Blue Waters Announcement: Winter Holiday services

  Created by: gbauer 2018-12-21 09:21:34 (Channels: systemnotices|general)

  Blue Waters partners;

  NCSA staff will observe the University of Illinois Winter break schedule starting 5pm Friday December 21, 2018, and will resume normal business hours on Wednesday, January 2nd, 2019. On-call staff will monitor Blue Waters and the service request system during this period, and respond to issues in a timely manner depending on the nature of the issue. Requests that require review by the project office such as reservations, quota increases or purge exemptions will be considered on Wednesday, January 2nd, 2019.

  Please send comments or concerns to help+bw@ncsa.illinois.edu.

  Enjoy the special job discount period that ends 1/2/2019: 50% charge discount on all queues, all jobs in projects that have consumed at least 25% of their current allocation.

  Season's Greetings from the Blue Waters team!

  
  Read Less

  Blue Waters partners; NCSA staff will observe the University of Illinois Winter break schedule starting 5pm Friday December 21, 2018, and will resume normal business hours on Wednesday, January 2nd, 2019. On-call staff will monitor Blue Waters and the service request system during this period, and respond to issues in a...
  Read More

• Blue Waters Return to Service

  Created by: tbouvet 2018-11-14 18:35:11 (Channels: unscheduledoutages)

  The meta data server failover on the Home file system completed at 6:23 PM CT. I/O transactions initiated during the failover should have resumed normal operation when failover completed. Blue Waters has resumed normal operations.

• Blue Waters File System Issue

  Created by: tbouvet 2018-11-14 17:45:36 (Channels: unscheduledoutages)

  Blue Waters is currently experiencing a meta data server failover on the home file system that began at 5:30 PM CT. As a result, all I/O transactions for the home filesystem will block until the failover completes on all clients. The rest of Blue Waters including the other filesystems are operating normally, but job scheduling has been paused. Recovery from this type event (without complications) usually takes less than 1 hour.

  
  Read Less

  Blue Waters is currently experiencing a meta data server failover on the home file system that began at 5:30 PM CT. As a result, all I/O transactions for the home filesystem will block until the failover completes on all clients. The rest of Blue Waters including the other filesystems are operating...
  Read More

• Blue Waters Return to Service

  Created by: tbouvet 2018-11-14 11:32:42 (Channels: unscheduledoutages)

  The meta data server failover on the Home file system completed at 11:26 AM CT. The file system issue start at 10:45 AM. I/O transactions initiated during the failover should have resumed normal operation when failover completed. Blue Waters has resumed normal operations.

• Blue Waters File System Issue

  Created by: tbouvet 2018-11-14 11:19:47 (Channels: unscheduledoutages)

  Blue Waters is currently experiencing a meta data server failover on the home file system that began at 11:15 AM CT. As a result, all I/O transactions for the home filesystem will block until the failover completes on all clients. The rest of Blue Waters including the other filesystems are operating normally, but job scheduling has been paused. Recovery from this type event (without complications) usually takes less than 1 hour.

  
  Read Less

  Blue Waters is currently experiencing a meta data server failover on the home file system that began at 11:15 AM CT. As a result, all I/O transactions for the home filesystem will block until the failover completes on all clients. The rest of Blue Waters including the other filesystems are operating...
  Read More

• Blue Waters Announcement: Ongoing Blue Waters Job Discounts

  Created by: gbauer 2018-11-12 21:44:41 (Channels: systemnotices|general|policychange)

  This is a reminder that the first period of job discounts ends on Thursday, November 15th. The second period starts on the 16th but with different criteria. Get your jobs submitted soon.

  Stop by the NCSA booth and say hi if you are at SC18.

  > The Blue Waters project is happy to announce two job discount periods as our way of saying thank you for another successful year of computing and the your help creating our 2017-2018 annual report.  

  > There are two discount periods:

  1. Nov 1 - Nov 15: 50% charge discount on all queues, all jobs completing during this period
  2. Nov 16 - Jan 2: 50% charge discount on all queues, all jobs completing during this period for projects that have consumed at least 25% of their current allocation.

  Please submit questions and comments to help+bw@ncsa.illinois.edu.

   

  
  Read Less

  This is a reminder that the first period of job discounts ends on Thursday, November 15th. The second period starts on the 16th but with different criteria. Get your jobs submitted soon. Stop by the NCSA booth and say hi if you are at SC18. > The Blue Waters project is happy to...
  Read More

• BlueWaters returned to service

  Created by: mshow 2018-11-05 06:18:32 (Channels: unscheduledoutages)

  The system has been returned to full service operation after a brief unscheduled outage.

• BlueWaters system issue

  Created by: mshow 2018-11-05 01:21:18 (Channels: unscheduledoutages|general)

  The system has experienced  a fault that will require a full system shutdown/reboot. This will take several hours before the system is returned to full operations. 

• Blue Waters Discount Update

  Created by: jenos 2018-11-02 17:22:33 (Channels: systemnotices|general|policychange)

  The previously announced discounts have not changed; however, they have not been activated yet to be reflected in current charges. This is expected to be activated next week, but will be retroactively applied to the original start date of Nov 1, 2018.

   

• Blue Waters Announcement: Blue Friday Job Discounts start today!

  Created by: jenos 2018-11-01 12:34:44 (Channels: systemnotices|general|policychange)

  The Blue Waters project is happy to announce two job discount periods as our way of saying thank you for another successful year of computing and the your help creating our 2017-2018 annual report.  

  There are two discount periods:

  1. Nov 1 - Nov 15: 50% charge discount on all queues, all jobs completing during this period
  2. Nov 16 - Jan 2: 50% charge discount on all queues, all jobs completing during this period for projects that have consumed at least 25% of their current allocation.

  Please submit questions and comments to help+bw@ncsa.illinois.edu.

   

  
  Read Less

  The Blue Waters project is happy to announce two job discount periods as our way of saying thank you for another successful year of computing and the your help creating our 2017-2018 annual report.   There are two discount periods: Nov 1 - Nov 15: 50% charge discount on all queues, all jobs completing during...
  Read More

• Blue Waters Notice: Temporary changes to shell stack size limit

  Created by: gbauer 2018-10-05 14:40:33 (Channels: systemnotices)

  Blue Waters partners,

  We would like to inform you of a change to the Blue Waters user environment that will be made today, Friday October 5th, at 2:30 pm Central time. A subsequent notice will be sent when the change are reverted. The change is:

  A. The shell stack size soft limit will be set to 2097152 kbytes and the hard limit will be set to 31000000 kbytes on login, MOM and compute nodes. 

  Please contact us at help+bw@ncsa.illinois.edu if your Blue Waters batch jobs are impacted by the stack size limit changes.

  We apologize for the inconvenience this might have.

  -Blue Waters Management

  
  Read Less

  Blue Waters partners, We would like to inform you of a change to the Blue Waters user environment that will be made today, Friday October 5th, at 2:30 pm Central time. A subsequent notice will be sent when the change are reverted. The change is: A. The shell stack size soft limit will be set to 2097152 kbytes and the...
  Read More

• Globus SSH-based CLI End of Life

  Created by: gbauer 2018-08-01 15:19:49 (Channels: systemnotices)

  As of August 1, 2018 the SSH-based command-line interface (CLI) to Globus is end-of-life.

  Please see view-source:https://bluewaters.ncsa.illinois.edu/data-transfer-doc (scroll to the CLI section) and/or the Globus CLI page https://docs.globus.org/cli/ for more information on the new Python based CLI.

  Send email to help+bw@ncsa.illinois.edu for assistance, questions, etc.

• Blue Waters Scheduled Maintenance: Return to Service

  Created by: gbauer 2018-07-16 19:43:38 (Channels: scheduledoutages|systemnotices|softwareupdate)

  Blue Waters Users,

  Blue Waters has been returned to service after a security update and programming environment update. CUDA 9.1 is now available as the default cudatoolkit but with gcc/4.9.3 remaining as the default GNU compiler. The default programming environment otherwise did not change to allow for an extended transition to the newer programming environment.

  For a complete discussion of the changes to the Programming Environment please see the Programming Environment (PE) Changes  portal page.

  This is an extensive programming environment update and as such there will be issues we did not discover in advance. Please report issues to help+bw@ncsa.illinois.edu.

  -Blue Waters Project

  ---

  A new Programming Environment was installed as a non-default and will be made default after a test period and subsequent announcement.

  CUDA notes: XK nodes will receive CUDA update (and associated driver) to version 9.1.  CUDA 7.5 will still be available as non-default.  CUDA 7.5 binaries have tested successfully in the CUDA 9.1 environment, but it may be necessary to recompile.  sm_20 and compute_20 are deprecated in any build or makefile setup.  People should use sm_35 and/or compute_35 to best match our Kepler K20x GPUs.

  
  Read Less

  Blue Waters Users, Blue Waters has been returned to service after a security update and programming environment update. CUDA 9.1 is now available as the default cudatoolkit but with gcc/4.9.3 remaining as the default GNU compiler. The default programming environment otherwise did not change to allow for an extended transition to...
  Read More

• Blue Waters Scheduled Maintenance Update

  Created by: jenos 2018-07-13 11:57:54 (Channels: scheduledoutages|systemnotices|softwareupdate)

  Blue Waters Users,

  Please note the following scheduled maintenance update:

  The scheduled maintenance for CUDA 9.1 deployment on Blue Waters has been expanded to a near full system service outage that will span 13 hours on Monday July 16th, from 9am to 10pm. This adjustment will eliminate the need for an additional scheduled maintenance period. To maximize availability, login nodes and filesystems will remain in service during the maintenance. The expanded activities include hardware maintenance, operating system software update roll ups, and some preventative maintenance to maximize future stability. Major changes are still the Programming Environment and CUDA 9.1 depoloyments originally planned.

   

  ---

  A new Programming Environment will be installed as a non-default and will be made default after a test period and subsequent announcement.

  CUDA notes: XK nodes will receive CUDA update (and associated driver) to version 9.1.  CUDA 7.5 will still be available as non-default.  CUDA 7.5 binaries have tested successfully in the CUDA 9.1 environment, but it may be necessary to recompile.  sm_20 and compute_20 are deprecated in any build or makefile setup.  People should use sm_35 and/or compute_35 to best match our Kepler K20x GPUs.

  Interim updates will be posted on the Blue Waters System News

   

  Please plan accordingly.

  Availability details: 

  Subsystem / Services	Projected Duration	Availability
  Compute	9am - 10pm	Unavailable
  Login / Filesystem		Available
  Scheduler	9am - 10pm	Unavailable
  Globus online endpoint (ncsa#BlueWaters)	9am - 10pm	Unavailable
  Globus online endpoint (ncsa#Nearline)	9am to 12pm (3 hours)	Unavailable

  
  Read Less

  Blue Waters Users, Please note the following scheduled maintenance update: The scheduled maintenance for CUDA 9.1 deployment on Blue Waters has been expanded to a near full system service outage that will span 13 hours on Monday July 16th, from 9am to 10pm. This adjustment will eliminate the need for an additional...
  Read More

• Blue Waters XK (GPU) resource and HPSS Nearline resources will be unavailable from 9am to 12pm on Monday July 16th

  Created by: kingda 2018-07-09 09:08:10 (Channels: scheduledoutages|systemnotices|softwareupdate)

  Blue Waters Users,

  Blue Waters XK compute resource and HPSS ncsa#Nearline resource will be undergoing maintenance Monday (July 16th) morning beginning at 9 AM. The XK compute nodes are anticipated to be unavailable for up to 3 hours. The login nodes and XE compute resource will remain in service.

  A new Programming Environment will be installed as a non-default and will be made default after a test period and subsequent announcement.

  CUDA notes: XK nodes will receive CUDA update (and associated driver) to version 9.1.  CUDA 7.5 will still be available as non-default.  CUDA 7.5 binaries have tested successfully in the CUDA 9.1 environment, but it may be necessary to recompile.  sm_20 and compute_20 are deprecated in any build or makefile setup.  People should use sm_35 and/or compute_35 to best match our Kepler K20x GPUs.

  Interim updates will be posted on the Blue Waters System News

   

  Please plan accordingly.

  Availability details: 

  Subsystem / Services	Projected Duration	Availability
  XK Compute	9am to 12pm (3 hours)	XK Unavailable
  XE Compute		Available
  Login / Filesystem		Available
  Scheduler		Available
  Globus online endpoint (ncsa#BlueWaters)		Available
  Globus online endpoint (ncsa#Nearline)	9am to 12pm (3 hours)	Unavailable

  
  Read Less

  Blue Waters Users, Blue Waters XK compute resource and HPSS ncsa#Nearline resource will be undergoing maintenance Monday (July 16th) morning beginning at 9 AM. The XK compute nodes are anticipated to be unavailable for up to 3 hours. The login nodes and XE compute resource will remain in service. A new Programming Environment will be installed...
  Read More

• Blue Waters Partial Scratch Outage

  Created by: tbouvet 2018-06-23 21:31:02 (Channels: unscheduledoutages|systemnotices)

  Blue Waters experienced a newtork switch failure that resulted in a partial outage of the scratch filesystem (ost168-179) from 7:44 PM CT to 7:59 PM CT. Jobs that ended during this time may have been impacted. I/O transactions targeting the affected storage server should block until the ost targets returned on all clients. Note: this brief outage was for a small portion of the scratch filesystem.

  
  Read Less

  Blue Waters experienced a newtork switch failure that resulted in a partial outage of the scratch filesystem (ost168-179) from 7:44 PM CT to 7:59 PM CT. Jobs that ended during this time may have been impacted. I/O transactions targeting the affected storage server should block until the ost targets returned...
  Read More

• Blue Waters Nearline Endpoint Busy

  Created by: gbauer 2018-06-21 23:07:24 (Channels: systemnotices)

  Due to very high demand for data retrieval from Nearline, a pause rule is in effect to allow manual task scheduling. You may submit tasks as normal and they will be run as quickly as possible.

  Tasks submitted to Globus will start in a paused state but will be released to run, at the earliest possible time, based on resource availability.

  We will update this posting as needed.

  We apologize for any inconvenience.

  Contact help+bw@ncsa.illinois.edu with questions or comments.

  
  Read Less

  Due to very high demand for data retrieval from Nearline, a pause rule is in effect to allow manual task scheduling. You may submit tasks as normal and they will be run as quickly as possible. Tasks submitted to Globus will start in a paused state but will be released to...
  Read More

• Blue Waters returned to full service

  Created by: mshow 2018-06-12 10:23:37 (Channels: mandatory|unscheduledoutages|systemnotices)

  Blue Waters has returned to full service after recovery from a power event.

• Blue Waters system power interruption

  Created by: mshow 2018-06-12 04:56:14 (Channels: mandatory|unscheduledoutages|systemnotices)

  Thunderstorms have resulted in a power interruption of the BlueWaters System. This outage imacts both the compute nodes and all  filesystems. Therefore, a full reboot will be necessary.Return to service is estimated to be approximately 10 am Centeral time.

  BW Admin

• Blue Waters has Returned to Service

  Created by: tbouvet 2018-06-07 14:27:12 (Channels: unscheduledoutages|systemnotices)

  Blue Waters has returned to full service at 2:14 PM CT. The issue encountered required a full system reboot to resolve. All running jobs were lost so please resubmit your jobs from latest checkpoint file if your job exited prematurely.   

• Blue Waters System Issue

  Created by: tbouvet 2018-06-07 10:12:41 (Channels: unscheduledoutages)

  Blue Waters is experiencing a full system issue that began at 6:30 AM CT. System support staff are evaluating and attempting to restore normal service but may require a full system reboot. Job scheduling is paused until the issue is resolved. Interim updates will be posted on the Blue Waters Message of the Day: https://bluewaters.ncsa.illinois.edu/motd

  UPDATE: 10AM
  Full System Reboot in progress. All running jobs were lost and will require a re-submission from latest checkpoint file. Lustre Filesystems, GO endpoints BlueWaters and Nearline will remain up, Login nodes will remain accessible.  

  
  Read Less

  Blue Waters is experiencing a full system issue that began at 6:30 AM CT. System support staff are evaluating and attempting to restore normal service but may require a full system reboot. Job scheduling is paused until the issue is resolved. Interim updates will be posted on the Blue Waters...
  Read More

• Blue Waters notice: Container software re-enabled on Blue Waters

  Created by: gbauer 2018-05-09 16:25:31 (Channels: systemnotices|softwareupdate)

  We are pleased to announce that the container software package Shifter and its associated service have been re-enabled on Blue Waters.

  Shifter was checked for vulnerabilities and none were found. To avoid concern, a patch was applied, and Shifter has been updated and re-enabled.

  Please see the Shifter page on the portal for updated documentation. 

  Please contact help+bw@ncsa.illinois.edu with your comments or questions.  

  
  Read Less

  We are pleased to announce that the container software package Shifter and its associated service have been re-enabled on Blue Waters. Shifter was checked for vulnerabilities and none were found. To avoid concern, a patch was applied, and Shifter has been updated and re-enabled. Please see the Shifter page on the portal for...
  Read More

• Blue Waters notice: Container software disabled on Blue Waters

  Created by: gbauer 2018-05-02 14:28:53 (Channels: systemnotices)

  This notice affects Blue Waters users relying on the container technologies of Shifter or Singularity. 

  Due to a recently announced security vulnerability related to container software, Shifter and Singularity container software have been disabled immediately to address the issue. We took this action to eliminate the risk of potential compromise of Blue Waters and related systems.  As this is a new issue, there is no ETA for a fix at this time, but we are pursuing a resolution and will notify further when container services can be restored. We are happy to work with you about possible mitigations for your work, but such mitigations will be specific to your workflows and use.

  We apologize for the inconvenience. Please check the Blue Waters the System News page for updates.

  Please contact help+bw@ncsa.illinois.edu with your comments or questions.  

  
  Read Less

  This notice affects Blue Waters users relying on the container technologies of Shifter or Singularity.  Due to a recently announced security vulnerability related to container software, Shifter and Singularity container software have been disabled immediately to address the issue. We took this action to eliminate the risk of potential compromise of Blue Waters...
  Read More

• Blue Waters Announcement: Winter Holiday services

  Created by: gbauer 2017-12-21 15:49:34 (Channels: systemnotices|general)

  Blue Waters partners;

  NCSA staff will observe the University of Illinois Winter break schedule starting 5pm Friday December 22, 2017, and will resume normal business hours on Tuesday, January 2nd, 2018. On-call staff will monitor Blue Waters and the service request system during this period and respond to issues in a timely manner depending on the nature of the issue. Requests that require review by the project office such as reservations, quota increases or purge exemptions will be considered on Tueday, January 2rd 2018.

   

  Please send comments or concerns to help+bw@ncsa.illinois.edu.

   

  Enjoy the special job discount period and happy holidays from the Blue Waters team!

   

   

  
  Read Less

  Blue Waters partners; NCSA staff will observe the University of Illinois Winter break schedule starting 5pm Friday December 22, 2017, and will resume normal business hours on Tuesday, January 2nd, 2018. On-call staff will monitor Blue Waters and the service request system during this period and respond to issues in a...
  Read More