Known issues: Difference between revisions

From Alliance Doc
Jump to navigation Jump to search
No edit summary
No edit summary
 
(90 intermediate revisions by 10 users not shown)
Line 1: Line 1:
<languages />
<languages />
<translate>
<translate>
= Intro = <!--T:1-->
* Please report issues to [mailto:support@computecanada.ca support@computecanada.ca]


= Shared issues = <!--T:2-->
<!--T:15-->
* The status page at http://status.computecanada.ca/ is not updated automatically yet, so may lag in showing current status.
Problems that affect many users and are being investigated, such as a cluster-wide malfunction or outage, are on [https://status.alliancecan.ca/ the Alliance Status page]. This "Known issues" page describes problems that affect many users but that may take some time to repair, or are not planned for repair at this time. Problems that only affect a specific software package are described on the wiki page for that software package.  
* Utilization accounting is not currently being forwarded to the [https://ccdb.computecanada.ca Compute Canada database]


== Scheduler errors == <!--T:6-->
==Report an issue== <!--T:1-->
* The CC slurm configuration preferentially encourages whole-node jobs. Users should, if appropriate, request whole-node rather than per-core resources. See [[Job_scheduling_policies#Whole_nodes_versus_cores;|Job Scheduling - Whole Node Scheduling]] ([[User:Pjmann|Patrick Mann]] 20:15, 17 July 2017 (UTC))
Please report issues to the [[Technical Support|technical support]] team.
* Cpu and Gpu backfill partitions have been created on both clusters. If a job is submitted with <24hr runtime, it will be automatically entered into the cluster-wide backfill partition. This partition has a low priority, but will allow increased utilization of the cluster by serial jobs. ([[User:Nathanw|Nathan Wielenga]])
* SLURM epilog does not fully clean up processes from ended jobs, especially if the job did not exit normally.    ([[User:Gbnewby|Greg Newby]]) Fri Jul 14 19:32:48 UTC 2017)
** This has been greatly improved after the addition of the epilog.clean script, but there are still nodes occasionally marked down for epilog failure. (NW)
* By default, the job receives environment settings from the submitting shell. This can lead to irreproducible results if it's not what you expect. To force the job to run with a fresh-like-login environment, you can submit with <tt>--export=none</tt> or add <tt>#SBATCH --export=NONE</tt> to your job script.


== Quota and filesystem problems == <!--T:7-->
==Shared issues== <!--T:2-->
=== Quota errors on /project filesystem ===
The [https://status.alliancecan.ca/ status page] is updated manually, so there may be a delay between when a problem begins and when it is posted to the status page.
Sometimes, users will see quota error on their project folders. This may happen when files are owned by a group other than the project group. You can change the group which owns files using the command
{{Command|chgrp -R <group> <folder>}}


<!--T:8-->
===Scheduler issues=== <!--T:6-->
To see what <group> should be, run the following command :
{{Command|stat -c %G $HOME/projects/*/}}


<!--T:9-->
<!--T:14-->
Only the owner of the files can run the <tt>chgrp</tt> command. To ask us to correct the group owner for many users, write to support@computecanada.ca
No known issues.


=== Nearline === <!--T:10-->
===Quota and filesystem issues=== <!--T:7-->
* "Nearline" capabilities are not yet available (see https://docs.computecanada.ca/wiki/National_Data_Cyberinfrastructure for a brief description of the intended functionality)
** Update July 17: still not working. If you need your nearline RAC2017 quota then please ask [mailto:support@computecanada.ca CC support]. ([[User:Pjmann|Patrick Mann]] 20:45, 17 July 2017 (UTC))


=== Missing symbolic links to project folders === <!--T:11-->
====Missing project folder==== <!--T:11-->
* Upon login to the new clusters, symbolic links are supposed to be created in the user's account, as described in [[Project layout]]. Sometimes, it does not happen. If this is the case, please verify that your access to the cluster is enabled on this page [https://ccdb.computecanada.ca/services/resources https://ccdb.computecanada.ca/services/resources]
Upon creation of a new account for a Principal Investigator, the [[Project layout|<code>/project</code>]] storage space might not be allocated until the next business day.


= Cedar only = <!--T:3-->
==Cluster-specific issues== <!--T:17-->
* SLURM operations will occasionally time out with a message like "Socket timed out on send/recv operation" or "Unable to contact slurm controller (connect failure)". As a temporary workaround, attempt to resubmit your jobs/commands, they should go through in a few seconds. ([[User:Nathanw|Nathan Wielenga]]) 08:50, 18 July 2017 (MDT))
** Should be resolved after a VHD migration to a new backend for slurmctl. (NW)
*Some people are getting an error "error: Job submit/allocate failed: Invalid account or account/partition combination specified"
**They need to specify '--account=<accounting group>'


= Graham only = <!--T:4-->
===Béluga=== <!--T:16-->
* /home is on an NFS appliance that does not support ACLs, so setfacl/getfacl doesn't work there.
No known issues.
** Workaround: use the /project or /scratch filesystems instead
** We're finding out whether this can be fixed through an update or reconfiguration.
* Compute nodes cannot access Internet
** Solution: Request exceptions to be made at support@computecanada.ca  Describe what you need to access and why.


<!--T:12-->
===Cedar=== <!--T:3-->
* Crontab is not offered on Graham. When attempting adding a new item there is an error during saving:
No known issues.
<pre>
 
[rozmanov@gra-login1 ~]$ crontab -e
===Graham=== <!--T:4-->
no crontab for rozmanov - using an empty one
Graham's /scratch is often slow; it will be replaced soon.
crontab: installing new crontab
 
/var/spool/cron/#tmp.gra-login1.XXXXKsp8LU: Read-only file system
===Narval=== <!--T:18-->
crontab: edits left in /tmp/crontab.u0ljzU
No known issues.
</pre>
Crontab does work on Cedar. So, there must be some kind of a common approach on CC system.
Clearly, the main issue is how to handle user's crontabs on multiple login nodes.
But it's not clear whether we even want to do so.


= Other issues = <!--T:5-->
#modules don't work for shells other than bash(sh)
#*Workaround (this appears to work but not tested extensively)
#**source $LMOD_PKG/init/tcsh
#**source $LMOD_PKG/init/zsh
#**source $LMOD_PKG/init/ksh
#*Bart will look at it in more detail
</translate>
</translate>

Latest revision as of 18:06, 25 October 2024

Other languages:

Problems that affect many users and are being investigated, such as a cluster-wide malfunction or outage, are on the Alliance Status page. This "Known issues" page describes problems that affect many users but that may take some time to repair, or are not planned for repair at this time. Problems that only affect a specific software package are described on the wiki page for that software package.

Report an issue

Please report issues to the technical support team.

Shared issues

The status page is updated manually, so there may be a delay between when a problem begins and when it is posted to the status page.

Scheduler issues

No known issues.

Quota and filesystem issues

Missing project folder

Upon creation of a new account for a Principal Investigator, the /project storage space might not be allocated until the next business day.

Cluster-specific issues

Béluga

No known issues.

Cedar

No known issues.

Graham

Graham's /scratch is often slow; it will be replaced soon.

Narval

No known issues.

Retrieved from "https://docs.alliancecan.ca/mediawiki/index.php?title=Known_issues&oldid=163369"