[
https://issues.apache.org/jira/browse/YARN-9370?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16793984#comment-16793984
]
Eric Yang commented on YARN-9370:
---------------------------------
The first line of logging seems to indicate a state of starting. There is no
logging for the ending state. It may concern people who is reading the log
file that trying to find the end state. Message variable might be best to
generate with StringBuilder or use slf4j API to improve logging performance:
{code}
LOG.info("ContainerId {} is assigned to GPU device {} on recovery.",
containerId, gpuDevice);
{code}
> Better logging in recoverAssignedGpus in class GpuResourceAllocator
> -------------------------------------------------------------------
>
> Key: YARN-9370
> URL: https://issues.apache.org/jira/browse/YARN-9370
> Project: Hadoop YARN
> Issue Type: Improvement
> Reporter: Szilard Nemeth
> Assignee: Yesha Vora
> Priority: Trivial
> Labels: newbie, newbie++
> Attachments: YARN-9370.001.patch
>
>
> The last line of
> org.apache.hadoop.yarn.server.nodemanager.containermanager.linux.resources.gpu.GpuResourceAllocator#recoverAssignedGpus
> is this:
> {code:java}
> usedDevices.put(gpuDevice, containerId);
> {code}
> We should have an info (or if not info, at least a debug) level log to
> indicate that a container is allocated to a GPU device during recovery.
> Please also check recovery related code, maybe there can be some room for
> improvement around logging.
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
---------------------------------------------------------------------
To unsubscribe, e-mail: yarn-issues-unsubscr...@hadoop.apache.org
For additional commands, e-mail: yarn-issues-h...@hadoop.apache.org
