Note: This bug is displayed in read-only format because
the product is no longer active in Red Hat Bugzilla.
Red Hat Satellite engineering is moving the tracking of its product development work on Satellite to Red Hat Jira (issues.redhat.com). If you're a Red Hat customer, please continue to file support cases via the Red Hat customer portal. If you're not, please head to the "Satellite project" in Red Hat Jira and file new tickets here. Individual Bugzilla bugs will be migrated starting at the end of May. If you cannot log in to RH Jira, please consult article #7032570. That failing, please send an e-mail to the RH Jira admins at rh-issues@redhat.com to troubleshoot your issue as a user management inquiry. The email creates a ServiceNow ticket with Red Hat. Individual Bugzilla bugs that are migrated will be moved to status "CLOSED", resolution "MIGRATED", and set with "MigratedToJIRA" in "Keywords". The link to the successor Jira issue will be found under "Links", have a little "two-footprint" icon next to it, and direct you to the "Satellite project" in Red Hat Jira (issue links are of type "https://issues.redhat.com/browse/SAT-XXXX", where "X" is a digit). This same link will be available in a blue banner at the top of the page informing you that that bug has been migrated.
DescriptionVedashree Deshpande
2022-03-31 14:15:39 UTC
Description of problem:
In satellite 6.10.4, while syncing any repository,
the error getting is:
[Errno 24] too many open files after some time syncing.
Version-Release number of selected component (if applicable):
6.10.4 although the customer says, he faces the issue in 6.9 as well as 6.10.3
Steps to Reproduce:
on Satellite 6.10.4 when syncing any repository.
Actual results:
[Errno 24] too many open files after some time syncing.
Expected results:
Repo should be synced.
Additional info:
Customer did try to fix it with the workaround which helped. But needs a fix or a deeper investigation.
~~~
checking the number of open files for the pulp user with
watch -n1 -t "lsof -u pulp -n | wc -l"
i reach over 4300 open files
which the pulp processes doent seem to be prepared to:
for i in $(ps waxu|grep ^pulp | awk '{print $2}') ; do prlimit --noheadings --pid $i --nofile; done
returns
1024:4096
manually setting it to a higher value with:
for i in $(ps waxu|grep ^pulp | awk '{print $2}') ; do prlimit --pid $i --nofile=4096:8192; done
~~~
While running a sync of RHEL 7 I am seeing it cap out around 1020 open files. I'm curious if this can still be produced on 6.10.6, and which repo specifically was being tested (I know it said "any" repo, but nonetheless)
We've fixed a few file leak bugs lately upstream, I believe at least some of those patches are already in 6.10.6.
Comment 20pulp-infra@redhat.com
2022-07-03 13:28:13 UTC
The Pulp upstream bug status is at closed. Updating the external tracker on this bug.
Comment 21pulp-infra@redhat.com
2022-07-03 13:28:16 UTC
All upstream Pulp bugs are at MODIFIED+. Moving this bug to POST.
Brian,
I am unable to reproduce any issues without an http proxy in the loop, and combined with the two reports and the traceback we can probably conclude that it is a necessary component of the bug.
I now have some results (which are attached) from using this command to sync the repo https://cdn.redhat.com/content/dist/rhel8/8/x86_64/baseos/os/ with no HTTP proxy using the following command:
watch -t -n 10 "(date '+TIME:%H:%M:%S' ; lsof -a -u pulp -n -d ^mem -d ^cwd -d ^rtd -d ^txt -d ^DEL | wc -l) | tee -a /tmp/pulp_nofile_sync"
Could you get results of the same command running through the duration of the sync, syncing the same repository, under an HTTP proxy, so that we can compare them?file:///home/dalley/devel/pulp_nofile_sync
Description of problem: In satellite 6.10.4, while syncing any repository, the error getting is: [Errno 24] too many open files after some time syncing. Version-Release number of selected component (if applicable): 6.10.4 although the customer says, he faces the issue in 6.9 as well as 6.10.3 Steps to Reproduce: on Satellite 6.10.4 when syncing any repository. Actual results: [Errno 24] too many open files after some time syncing. Expected results: Repo should be synced. Additional info: Customer did try to fix it with the workaround which helped. But needs a fix or a deeper investigation. ~~~ checking the number of open files for the pulp user with watch -n1 -t "lsof -u pulp -n | wc -l" i reach over 4300 open files which the pulp processes doent seem to be prepared to: for i in $(ps waxu|grep ^pulp | awk '{print $2}') ; do prlimit --noheadings --pid $i --nofile; done returns 1024:4096 manually setting it to a higher value with: for i in $(ps waxu|grep ^pulp | awk '{print $2}') ; do prlimit --pid $i --nofile=4096:8192; done ~~~