Oh no, do not just touch(1) an RPM metadata file, that way lies the lair of dark Chtulu, the very sight of whom induces madness... (Source: tried that once, oh dear god what a mess it made until I nuked my entire rpm db from orbit.)
But I think you can clean one repo at a time. And/or *deleting* the repo metadata file should work, too.
-Adam
Get Outlook for Androidhttps://aka.ms/AAb9ysg ________________________________ From: Roundtable roundtable-bounces@muug.ca on behalf of Trevor Cordes trevor@tecnopolis.ca Sent: Tuesday, November 29, 2022 8:20:29 PM To: roundtable@muug.ca roundtable@muug.ca Subject: Re: [RndTbl] Remi repo bug?
On 2022-11-29 Gilbert Detillieux wrote:
I’m surprised that you haven’t seen that before, since it’s a fairly common occurrence on RHEL7 systems. I’m also surprised a net search also came up empty... Maybe EL7 is just getting too old to be relevant.
Anyway this is the command you need:
yum clean all
Might be good to do before the update runs in cron, if this comes up again.
Ya, but if you run that every hour then yum will need to d/l the full 100MB+ of metadata files every hour? I guess it's RS's problem, not mine, but still...
On 2022-11-29 Adam Thompson wrote:
That's a problem with a Remi mirror site. What hostname are you pointed at? -Adam
mirrorlist=http://cdn.remirepo.net/enterprise/7/safe/mirror
Looks like it'll do the auto-mirror-selection thing.
I was thinking maybe instead of a yum clean I could do a touch on the file it's complaining about to make it "older", then yum will just redownload *just* Remi.
_______________________________________________ Roundtable mailing list Roundtable@muug.ca https://muug.ca/mailman/listinfo/roundtable