Bug 470862
| Summary: | condor_schedd log spam from chown of removed directories | ||
|---|---|---|---|
| Product: | Red Hat Enterprise MRG | Reporter: | Matthew Farrellee <matt> |
| Component: | grid | Assignee: | Matthew Farrellee <matt> |
| Status: | CLOSED ERRATA | QA Contact: | Kim van der Riet <kim.vdriet> |
| Severity: | medium | Docs Contact: | |
| Priority: | medium | ||
| Version: | 1.0 | ||
| Target Milestone: | 1.1 | ||
| Target Release: | --- | ||
| Hardware: | All | ||
| OS: | Linux | ||
| Whiteboard: | |||
| Fixed In Version: | Doc Type: | Bug Fix | |
| Doc Text: | Story Points: | --- | |
| Clone Of: | Environment: | ||
| Last Closed: | 2009-02-04 16:05:27 UTC | Type: | --- |
| Regression: | --- | Mount Type: | --- |
| Documentation: | --- | CRM: | |
| Verified Versions: | Category: | --- | |
| oVirt Team: | --- | RHEL 7.3 requirements from Atomic Host: | |
| Cloudforms Team: | --- | Target Upstream Version: | |
| Embargoed: | |||
Fixed upstream, will be in 7.1.4-0.5
--
commit cf3b59a9120ddce221479c578af218ca313c3072
Author: Matthew Farrellee <matt>
Date: Tue Nov 11 13:03:10 2008 -0600
dprintf at D_FULLDEBUG (not D_ALWAS) when "...User may run into permissions problems..." for now, proper solution uses CondorError stack
src/condor_schedd.V6/schedd.C | 2 +-
1 files changed, 1 insertions(+), 1 deletions(-)
commit 231e9ca234ef9f427a827da5e3fbe6d6fa1a6e9d
Author: Matthew Farrellee <matt>
Date: Tue Nov 11 12:50:54 2008 -0600
dprintf at D_FULLDEBUG (not D_ALWAYS) when "Error: Unable to chown", which happens, and is acceptable, in common cases (e.g. no spool dir)
src/condor_c++_util/directory.C | 2 +-
1 files changed, 1 insertions(+), 1 deletions(-)
commit 896110389180ae0ce3a17cb1ec62a487d8b82736
Author: Matthew Farrellee <matt>
Date: Tue Nov 11 12:48:34 2008 -0600
dprintf at D_FULLDEBUG (not D_ALWAYS) about "Attempting to chown" a file that does not exist
src/condor_c++_util/directory.C | 2 +-
1 files changed, 1 insertions(+), 1 deletions(-)
An advisory has been issued which should help the problem described in this bug report. This report is therefore being closed with a resolution of ERRATA. For more information on therefore solution and/or where to find the updated files, please follow the link below. You may reopen this bug report if the solution does not work for you. http://rhn.redhat.com/errata/RHBA-2009-0036.html |
Description of problem: The condor_schedd logs failures to chown non-existent job spool directories at level D_ALWAYS. Version-Release number of selected component (if applicable): 7.1.4-0.4 (likely present in 7.0.5 and earlier) How reproducible: Always Steps to Reproduce: 0. condor must be started by root 1. start watching SchedLog 2. submit a job 3. remove the job 4. in SchedLog, see: Attempting to chown '...', but it doesn't appear to exist. Error: Unable to chown '...' from X to Y.Y Failed to chown ... from X to Y.Y. User may run into permissions problems when fetching sandbox. Expected results: I'd expect the condor_schedd not to be so verbose about an issue that appears to be harmless. Additional info: Why is this a problem? Submit and remove 1000 jobs and watch the condor_schedd waste large amounts of time logging and the SchedLog rotate over and over. The lower levels of the code should return errors and the higher levels can determine how and what to log. The lower levels shouldn't log above D_FULLDEBUG. This only happens when condor is run by root, when run by a un-privileged user it does not try to chown anything.