Red Hat Bugzilla – Bug 1441041
[RFE] - Include imageio/daemon.log and ovirt-imageio-proxy/image-proxy.log in LC/sos output
Last modified: 2018-04-10 14:03:36 EDT
1. Proposed title of this feature request Include imageio/daemon.log and ovirt-imageio-proxy/image-proxy.log in Log Collector (sosreport) output 3. What is the nature and description of the request? Need to include imageio/daemon.log and ovirt-imageio-proxy/image-proxy.log in Log Collector (sosreport) output 4. Why does the customer need this? For customers having trouble with the disk upload feature, these logs are necessary for further troubleshooting. Not including these logs be default causes unnecessary back-and-forth 5. How would the customer like to achieve this? (List the functional requirements here) Relevant plugins need to be modified or created: rhevm: /var/log/ovirt-imageio-proxy/image-proxy.log* host/hypervisors: /var/log/ovirt-imageio-daemon/daemon.log* 6. For each functional requirement listed, specify how Red Hat and the customer can test to confirm the requirement is successfully implemented. - generate Log Collector - confirm above files are included 7. Is there already an existing RFE upstream or in Red Hat Bugzilla? No 8. Does the customer have any specific timeline dependencies and which release would they like to target (i.e. RHEL5, RHEL6)? No 9. Is the sales team involved in this request and do they have any additional input? No 10. List any affected packages or components. sos (RHEL7 repo) ovirt-log-collector (rhel-7-server-rhv-4.0-rpms repo) 11. Would the customer be able to assist in testing this functionality if implemented? It's easy to test, but I am happy to assist if required.
Not sure if we should create another plugin or modify and existing one. Currently for ovirt: # grep var.log ovirt*py ovirt_hosted_engine.py: SETUP_LOG_GLOB = '/var/log/ovirt-hosted-engine-setup/*.log' ovirt_hosted_engine.py: HA_LOG_GLOB = '/var/log/ovirt-hosted-engine-ha/*.log' ovirt_hosted_engine.py: '/var/log/ovirt-hosted-engine-ha/agent.log', ovirt_hosted_engine.py: '/var/log/ovirt-hosted-engine-ha/broker.log', ovirt.py: # backtrace written to '/var/log/ovirt-engine/console.log ovirt.py: "/var/log/ovirt-engine", ovirt.py: "/var/log/ovirt-engine-dwh", ovirt.py: "/var/log/ovirt-engine-reports", ovirt.py: "/var/log/ovirt-scheduler-proxy", ovirt.py: "/var/log/rhevm", ovirt.py: "/var/log/ovirt-guest-agent", and VDSM: # grep var.log vdsm*py self.__addCopySpecLogLimit("/var/log/vdsm/*", logsize) self.__addCopySpecLogLimit("/var/log/vdsm-reg/*", logsize) self.addCopySpec("/var/log/ovirt.log") self.addCopySpec("/var/log/sanlock.log") self.collectExtOutput("/bin/ls -l /var/log/core")
Re-scheduling for potential inclusion in 7.5 (7.4 is in testing phase already). Some questions: 1) How to detect a system that is expected to have those logfiles? (i.e. if I want to collect /var/log/httpd, I can do so if and only if httpd package is installed)? What package presence on the system shall trigger collecting either of the logs? 2) Shall the logs be collected completed, or just until some log size limit (in case there are plenty of older logs, shall we skip collecting the oldest?)? 3) Can't be there some secret (password, private key,..) in the logs? I guess no and I dont require detailed analysis, since logs usually dont contain customer sensitive information, but if it is known there is something to obfuscate / what customers wont like to share, please describe.
> 1) How to detect a system that is expected to have those logfiles? for host/hypervisor - ovirt-imageio-daemon for rhevm - ovirt-imageio-proxy > 2) Shall the logs be collected completed, or just until some log size limit The logs aren't too verbose, so getting everything should be fine > 3) Can't be there some secret (password, private key,..) in the logs? I can't see any username/passwords, nor anything that would not be included in Log Collector/sosreport anyway
I dont think this shall be glued into either "base" ovirt plugin or ovirt_hosted_engine or even libvirt/virsh/.. plugins, sicne my understanding is ovirt imageio is an optional part of ovirt - am I right? So I would suggest creating a new plug-in that will be enabled by presence of _either_ ovirt-imageio-daemon or ovirt-imageio-proxy packages, and will collect both logfiles (it is expected just one such logfile to exist but that's no issue for sos - it does not attempt to collect non-existing files). Is there some command output or a config file worth to gather as well by the new plugin?
Hi Pavel, I believe, it's optional. Yes I agree with with your suggestion. Additional config files worth grabbing: /etc/ovirt-imageio-daemon/logger.conf [ from the daemon ] /etc/ovirt-imageio-proxy/ovirt-imageio-proxy.conf [ from the proxy ]
Upstream PR raised. Could you please check if neither config files the new plug-in collects can contain any sensitive information we shall obfuscate?
Hi Pavel, I checked my environment and can't see anything sensitive.
Steve, could you please pm_ack+ this Futurefeature? It adds a new sos plugin for oVirt Image I/O.
modified due to rebase of sos to 3.5
Added the plugin to log-collector in https://gerrit.ovirt.org/#/c/84596/
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-2018:0963