Bug 1699381
| Summary: | [RFE] sos for the gluster commands | ||
|---|---|---|---|
| Product: | Red Hat Enterprise Linux 7 | Reporter: | Bhushan Ranpise <branpise> |
| Component: | sos | Assignee: | Pavel Moravec <pmoravec> |
| Status: | CLOSED ERRATA | QA Contact: | Miroslav Hradílek <mhradile> |
| Severity: | medium | Docs Contact: | |
| Priority: | unspecified | ||
| Version: | 7.7 | CC: | agk, bmr, jhunsaker, mhradile, myllynen, plambri, pmoravec, sasundar, sbradley |
| Target Milestone: | rc | Keywords: | FutureFeature, OtherQA |
| Target Release: | --- | ||
| Hardware: | x86_64 | ||
| OS: | Linux | ||
| Whiteboard: | |||
| Fixed In Version: | sos-3.8-3.el7 | Doc Type: | If docs needed, set a value |
| Doc Text: | Story Points: | --- | |
| Clone Of: | Environment: | ||
| Last Closed: | 2020-03-31 20:04:09 UTC | Type: | Bug |
| Regression: | --- | Mount Type: | --- |
| Documentation: | --- | CRM: | |
| Verified Versions: | Category: | --- | |
| oVirt Team: | --- | RHEL 7.3 requirements from Atomic Host: | |
| Cloudforms Team: | --- | Target Upstream Version: | |
| Embargoed: | |||
Following up on #c5:
- new to add:
gluster volume get <volname> all (for all volumes)
gluster get-state
killall -USR1 glusterd (to be added to glusterfs/glusterfsd killall cmd, under dump option)
for gpid in $(ps aux | grep "glusterfs " | grep :_ | awk '{print $2}') ; do kill -USR1 $gpid ; done (also under dump option)
ps -Ll -p <pid of a zombie process>
cat /proc/<pid of a zombie process>/task/*/stack
I recall there were some concerns about /proc collection, but havent found anything specific to tasks/stack - this seem to me safe.
Question1: Shall we do so for *all* zombie processes, even unrelated to gluster? Or how to filter what zombie processes to get details for?
Question2: We can also enable the '-k gluster.dump=on' via some new RHHI preset in sosreport - just let me know how to determine the system belongs to RHHI (i.e. some specific package installed?).
Further, how much sense it has to take /proc/<pid of a zombie process>/task/*/stack as a zombie process is just a PID (so any stack must be empty)? > Question2: We can also enable the '-k gluster.dump=on' via some new RHHI > preset in sosreport - just let me know how to determine the system belongs > to RHHI (i.e. some specific package installed?). A preset could work - from BZ1699401 we'll be adding a RHHI profile to sos-collector, so we can get it from both directions here. RHHI requires the presence of _both_ the gluster and rhv packages. For the nodes that means the qemu-kvm-rhev package. Jake, could you pls. verify this BZ? A yum repository for the build of sos-3.8-1.el7 (task 23200819) is available at: http://brew-task-repos.usersys.redhat.com/repos/official/sos/3.8/1.el7/ You can install the rpms locally by putting this .repo file in your /etc/yum.repos.d/ directory: http://brew-task-repos.usersys.redhat.com/repos/official/sos/3.8/1.el7/sos-3.8-1.el7.repo RPMs and build logs can be found in the following locations: http://brew-task-repos.usersys.redhat.com/repos/official/sos/3.8/1.el7/noarch/ The full list of available rpms is: http://brew-task-repos.usersys.redhat.com/repos/official/sos/3.8/1.el7/noarch/sos-3.8-1.el7.src.rpm http://brew-task-repos.usersys.redhat.com/repos/official/sos/3.8/1.el7/noarch/sos-3.8-1.el7.noarch.rpm Build output will be available for the next 21 days. Hello, could you please verify this BZ? Thanks in advance. A yum repository for the build of sos-3.8-3.el7 (task 23918180) is available at: http://brew-task-repos.usersys.redhat.com/repos/official/sos/3.8/3.el7/ You can install the rpms locally by putting this .repo file in your /etc/yum.repos.d/ directory: http://brew-task-repos.usersys.redhat.com/repos/official/sos/3.8/3.el7/sos-3.8-3.el7.repo RPMs and build logs can be found in the following locations: http://brew-task-repos.usersys.redhat.com/repos/official/sos/3.8/3.el7/noarch/ The full list of available rpms is: http://brew-task-repos.usersys.redhat.com/repos/official/sos/3.8/3.el7/noarch/sos-3.8-3.el7.src.rpm http://brew-task-repos.usersys.redhat.com/repos/official/sos/3.8/3.el7/noarch/sos-3.8-3.el7.noarch.rpm Build output will be available for the next 21 days. (the independent (/var)/run : will folowup on it later). Since the problem described in this bug report should be resolved in a recent advisory, it has been closed with a resolution of ERRATA. For information on the advisory, and where to find the updated files, follow the link below. If the solution does not work for you, open a new bug report. https://access.redhat.com/errata/RHEA-2020:1127 |
Description of problem: I have a case- 02358588, where the customer wants the gluster-related commands to be captured by the gluster sos plugin itself: - gluster vol info - gluster vol status - gluster peer status - gluster volume get <volname> all (for all volumes) - gluster get-state - "killall -USR1 glusterd" on the node where you see the problem and in one node without it to compare to see glusterd state info - "for gpid in $(ps aux | grep "glusterfs " | grep :_ | awk '{print $2}') ; do kill -USR1 $gpid ; done" on node with hanging VM to collect glusterfs state info - ps -Ll -p <pid of a zombie process> - cat /proc/<pid of a zombie process>/task/*/stack Version-Release number of selected component (if applicable): How reproducible: Steps to Reproduce: 1. 2. 3. Actual results: Expected results: Additional info: