To summarise, yes, I freed up some space and then restarted all services, and no, I didn't have to do anything in Mongodb. I'm not sure that I needed to restart all of the services, but I was in a hurry to just get things working. As far as helping out, I might volunteer to help out once I know more about Pulp but I'm still a complete newbie and have to get it all set up and working here first.
On 20 March 2015 at 13:58, Michael Hrivnak <[email protected]> wrote: > Alex, > > I'm glad you were able to get things back up and running. So would you > summarize your path to success as 1) free up some space, 2) restart > services, and then everything was fine? Did you have to do anything with > mongodb? > > FWIW, my favorite way to quickly free up space and inodes is "yum clean > all". > > There was discussion of this recently, that we need to investigate all the > ways that pulp could fail when the filesystem fills up. Presumably most of > that would happen during a sync process, and ideally pulp would be able to > gracefully stop what it's doing and fail the task. > > If someone, perhaps yourself, is willing to spend a little time > experimenting with causing a disk-full situation and collecting info about > how pulp behaves, and where it got stuck, that would be very helpful! I and > others would be happy to brainstorm on how to setup a VM for this, what > data to collect, etc. > > Michael > > ----- Original Message ----- > From: "Alex Harvey" <[email protected]> > To: [email protected] > Sent: Thursday, March 19, 2015 9:57:52 PM > Subject: Re: [Pulp-list] recover from filesystem full > > OK - I got past this by deleting the RPMs and then restarting all the > services. Still, it would be good to know how to recover from a full > filesystem in general, without deleting RPMs and then deleting a repository. > > On 20 March 2015 at 12:03, Alex Harvey < [email protected] > wrote: > > > > Hi all, > > I've filled up the filesystem in /var/lib/pulp. I can't find any > instructions on how to recover from this situation. I tried just deleting > all the RPMs that I had downloaded but there's evidently some corruption > relating to the filesystem filling. > > I just see this > > > > # pulp-admin -u admin -p admin rpm repo delete --repo-id centos-6.5-x86_64 > > This command may be exited via ctrl+c without affecting the request. > > [-] > > > Running... > > [|] > > Waiting to begin... > > Hangs forever. Logs say > > > > # tail -0f /var/log/messages > > Mar 20 10:25:16 localhost pulp: celery.worker.strategy:INFO: Received > task: > pulp.server.async.tasks._reserve_resource[cd24998e-c57a-43be-bcea-a4b4f2f6410c] > > Mar 20 10:25:16 localhost pulp: celery.worker.job:INFO: Task > pulp.server.async.tasks._reserve_resource[cd24998e-c57a-43be-bcea-a4b4f2f6410c] > succeeded in 0.047468271805s: > u'[email protected]' > > Mar 20 10:25:16 localhost pulp: celery.worker.strategy:INFO: Received > task: > pulp.server.tasks.repository.delete[eb2c073f-69f9-4b77-b3fc-cff12e936a7d] > Best regards > Alex > > > _______________________________________________ > Pulp-list mailing list > [email protected] > https://www.redhat.com/mailman/listinfo/pulp-list >
_______________________________________________ Pulp-list mailing list [email protected] https://www.redhat.com/mailman/listinfo/pulp-list
