File locking issues causing access to M: and S: drives
Update - Yesterday's errant file locks were successfully cleared at 5am this morning, fyi.
Sep 16, 2026 - 09:15 NZST
Update - We are confirming that the daily restart of the protocol nodes was successfully suspended over the weekend. A restart occurred at 0500 Friday, then again at 0500 Monday.
Sep 09, 2026 - 09:51 NZST
Update - At the request of BSI, the 0500hrs daily restart of the protocol nodes has been modified to only happen on business days, Mon - Fri. This to allow a particular workstream to run for more than 24hrs at the weekends. Thus, errant file locks that occur on Fridays will now be cleared on Monday mornings at 0500hrs.
Sep 03, 2026 - 13:33 NZST
Update - Yesterday's errant file locks were successfully cleared at 5am this morning, fyi.
Sep 02, 2026 - 08:35 NZST
Update - Yesterday's errant file locks were successfully cleared at 5am this morning, fyi.
Aug 28, 2026 - 08:59 NZST
Update - The restart of the protocol node services at 0500hrs this morning has been verified as successful. All file locks have been cleared. The M: and S: drives, and any NFS mounts, are available
Aug 20, 2026 - 05:20 NZST
Investigating - We are acknowledging there has been considerable recent disruption due to the Samba file locking issue and access to the M: and S: drives. We apologise for this disruption and are working hard on finding a solution or at least a workaround to mitigate the impact.
The first mitigation will be an automated nightly restart of the protocol nodes which serve these Windows drives. This will occur at 0500hrs daily with an outage of 15 minutes each time. The first of these automated restarts will be scheduled for 0500 Thursday Aug 20th.
Further to that we are working on a POC that involves redesigning how the Samba service is delivered. This first needs building and testing to ensure it alleviates the problem, but we are addressing this with urgency. We will provide further updates on this approach when appropriate.
Aug 18, 2026 - 16:36 NZST

About This Site

Bioeconomy Science Institute - AgResearch eRI status

Uptime over the past 90 days. View historical uptime.
Identity Broker Service Operational
Managed Storage Service Degraded Performance
General Flexi HPC Platform Operational
Network connectivity Operational
Compute cluster Operational
Login nodes Operational
Operational
Degraded Performance
Partial Outage
Major Outage
Maintenance
Major outage
Partial outage
No downtime recorded on this day.
No data exists for this day.
had a major outage.
had a partial outage.

No incidents or maintenance related to this downtime.

Sep 16, 2026

User Support System Maintenance
Resolved - The BSI portal can now also be used to log tickets, https://reannz.atlassian.net/servicedesk/customer/portal/106/group/123. We only have a few minor issues left with Jira, not specifically related to BSI. So this incident is resolved.
Sep 16, 09:14 NZST
Update - Our Jira problems are largely resolved now. We have confirmed users can now successfully use support@nesi.org.nz to log new tickets. But do let us know if you don't receive a reply or your email gets bounced.
Sep 8, 15:54 NZST
Update - We are aware that some users continue to have issues logging support tickets and we are fixing these as we identify them. However, once again, if you are not receiving responses from our Jira system in a timely manner then please email me directly at greg.hall@reannz.co.nz with your issue. I will forward these to a Support team member as appropriate.
Sep 1, 11:42 NZST
Update - Jira is in better shape but we continue to work through some issues. There is a new customer portal that can be used for logging Support tickets at https://reannz.atlassian.net/servicedesk/customer/portal/106/group/123 or alternatively just email support@nesi.org.nz.
Please use these methods for logging tickets, but some users may not be receiving our outbound responses. If you have not received a response in a timely manner you may still email me directly with your ticket number, and I will follow up, greg.hall@reannz.co.nz
Aug 28, 12:21 NZST
Monitoring - We are continuing to have some issues with our Jira/JSM ticketing system, work with our third party vendors continues. However we are able to receive and interact with tickets sent to support@nesi.org.nz, so please continue using this interface.
If you are not getting any response and need to escalate please contact me directly at greg.hall@reannz.co.nz
Aug 27, 09:39 NZST
Update - We are in an improved state although out third party vendors are continuing to provide support. However we are advanced enough now to revert to using the support@nesi.org.nz email for creating tickets.

Further updates tomorrow morning
Aug 26, 16:24 NZST
Update - In the meantime we continue to monitor the health of the eRI platform. However if you have an urgent P1 or P2 issue please email me directly at greg.hall@reannz.co.nz
Aug 26, 11:02 NZST
Investigating - We have been performing scheduled maintenance of our User Support systems, Jira, overnight and have encountered some unexpected issues. Our third party providers are working to resolve the issues. Consequently my.nesi.org .nz is unavailable, and emails sent to support@nesi.org.nz are being queued but are not currently visible to us.
We expect these services to be restored by 5:00 pm Wednesday 26 August. Any emails sent to support@nesi.org.nz during the maintenance window will be responded to once services are restored.
Aug 26, 10:49 NZST
compute-1 is currently under maintenance
Resolved - compute-1 was re-instated to the cluster yesterday afternoon. Unfortunately the maintenance was not completed. This has been rescheduled for Monday Sep 21st. A maintenance notice will be posted accordingly
Sep 16, 08:51 NZST
Identified - compute-1 has been reporting DIMM failures. It has been drained and put into maintenance mode whilst this h/w fault is addressed. We will also take the opportunity to apply a BIOS upgrade. The node should be available by the end of today.
Sep 15, 14:04 NZST

Sep 15, 2026

Sep 14, 2026

No incidents reported.

Sep 13, 2026

No incidents reported.

Sep 12, 2026

No incidents reported.

Sep 11, 2026

No incidents reported.

Sep 10, 2026

login-0 outage for a hypervisor upgrade, Thursdsay morning
Completed - this is completed
Sep 10, 08:59 NZST
Verifying - login-0 and interactive-1 have been successfully migrated, and GPFS restarted. The filesystems are available, Nix has been tested ok. Both systems are now available.
Sep 3, 09:22 NZST
In progress - Scheduled maintenance is currently in progress. We will provide updates as necessary.
Sep 3, 09:00 NZST
Scheduled - We will be migrating login-0 and interactive-1 on Thursday, Sep 3rd from 0900hrs. This will require a 30 minute outage of login-0 whilst we restart GPFS there. Any sessions/processes requiring access to the GPFS file systems on login-0 will die. During this outage login-1 will remain available.
interactive-1 will also be affected. It will be drained from Slurm, GPFS restarted, then resumed in Slurm during the same window.
Sep 1, 13:17 NZST

Sep 9, 2026

Sep 8, 2026

Sep 7, 2026

No incidents reported.

Sep 6, 2026

No incidents reported.

Sep 5, 2026

No incidents reported.

Sep 4, 2026

No incidents reported.

Sep 3, 2026

Duplicate network packets occurring on the idm1 Identity Management server
Resolved - Confirmed this morning there are no duplicated packet issues on idm1. This issue is resolved.
Sep 3, 09:23 NZST
Monitoring - A network switch change has been made and the duplicated packet issues has been resolved. We will continue to monitor
Sep 2, 16:08 NZST
Investigating - We have discovered a networking issue affecting the Identity server idm1. This is not a hard fault but may be contributing to the slow performance issues we have recently been advised of. We are actively investigating and will provide updates when we know more.
Sep 2, 11:57 NZST
gquery module load failures
Resolved - Gquery testing completed successfully this morning
Sep 3, 09:12 NZST
Monitoring - We will continue monitoring but testing has been successful
Sep 2, 11:23 NZST
Update - We have restarted a service on both login nodes and have successfully run "module load gquery" on both nodes. Please test gquery further and advise
Sep 2, 10:56 NZST
Investigating - We are investigating this issue with urgency. We have found "module load gquery" is working on login-0 but fails on login-1
Sep 2, 10:12 NZST

Sep 2, 2026

Hypervisor OS and firmware upgrade - no impact.
Completed - The migrations were completed by about 1000hrs this morning with no impact
Sep 2, 16:09 NZST
In progress - Scheduled maintenance is currently in progress. We will provide updates as necessary.
Sep 2, 09:00 NZST
Scheduled - We will be live migrating the Coldfront instance, and several gquery instances from the hypervisor in question. We have successfully live migrated these instances in the past and don't expect users to be impacted. The migrations will take less than 30 minutes from 0900hrs Wednesday.
Aug 31, 16:48 NZST
Incident History Powered by Atlassian Statuspage