Bug 1331870
| Summary: | Looks like /etc/updatedb.conf does have a PRUNEFS variable updated with fuse.glusterfs | ||
|---|---|---|---|
| Product: | Red Hat Enterprise Linux 7 | Reporter: | Peter Portante <pportant> |
| Component: | mlocate | Assignee: | Michal Sekletar <msekleta> |
| Status: | CLOSED ERRATA | QA Contact: | Karel Srot <ksrot> |
| Severity: | medium | Docs Contact: | |
| Priority: | high | ||
| Version: | 7.1 | CC: | dominique.martinet, ksrot, ndevos, pportant, rcyriac, smohan |
| Target Milestone: | pre-dev-freeze | Keywords: | FastFix |
| Target Release: | 7.5 | ||
| Hardware: | All | ||
| OS: | Linux | ||
| Whiteboard: | |||
| Fixed In Version: | mlocate-0.26-8.el7 | Doc Type: | Bug Fix |
| Doc Text: | Story Points: | --- | |
| Clone Of: | Environment: | ||
| Last Closed: | 2018-04-10 18:45:34 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: | |||
| Bug Depends On: | |||
| Bug Blocks: | 1465901, 1469361 | ||
|
Description
Peter Portante
2016-04-29 21:03:59 UTC
Just like it is done for other filesystems, I prefer to see fuse.glusterfs added to the default PRUNEFS variable in the updatedb.conf from the mlocate package. Could you move this bug to the RHEL version that you are interested in, and maybe clone it for others? Bug 1331944 has been filed for Fedora, the distributions default updatedb.conf file is located here: http://pkgs.fedoraproject.org/cgit/rpms/mlocate.git/log/updatedb.conf Seems buggy to pass this to a central package. Why can't GlusterFS update the configuration to make sure it is good for their users? Especially in our products? Our products include mlocate. Part of the productization is integrating with other components. Gluster should try hard not modify configuration files that other packages ship, and are not intended for automated modifications. If mlocate would offer a directory where Gluster can drop-in a config file, things would be different. We do not know how an administration modifies existing configuration files, and breaking that is bad behaviour. Would you be opposed to it when I move this bug to RHEL/mlocate as suggested in comment #2? So /etc/updatedb.conf is supposed to be updated to add file systems and file system types that should not be considered, right? What marks a file as not intended for automatic update? It would be nice to have mlocate read a directory of config files to build up its notion of state. But since we don't have that, wouldn't it be our place to make our product take this on until either upstream handles it properly? This bug nearly took out my system from service by consuming the system disk. That did not happen until Gluster was installed. So, it seems like GlusterFS as a product should own this until some other part of the system does. I will not oppose moving this to RHEL/mlocate, as I don't think that is my place. It would nice to see that we can own the problems sufficiently to make sure they are solved for customers before handing them off to another team. Moving this to RHEL/mlocate.
Because of this description, I think that a "high" priority is in order:
> This bug nearly took out my system from service by consuming the system
> disk. That did not happen until Gluster was installed. So, it seems like
> GlusterFS as a product should own this until some other part of the system
> does.
It would be an ugly workaround to try and modify a configuration file through some Gluster related scripts. There is no guarantee that the configuration file is available during the installation of the glusterfs-fuse package, mlocate can get installed afterwards. We would only be addressing one option, and may require more workarounds to address other cases.
Adding "fuse.glusterfs" to the PRUNFE variable in /etc/updatedb.conf that comes with mlocate is the best approach IMHO.
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/RHBA-2018:1004 |