I've been seeing the below message from yum whenever the repo has an update (CentOS 5):
/etc/cron.daily/yum.cron:
** Message: sqlite cache needs updating, reading in metadata
Googling a bit, it looks like others have seen this happen as well. The solutions, when I've found them, have been along the lines of "send output to /dev/null" or "edit this file", but nothing that feels like the "right" fix.
Is there a general solution to this? Or should we just sit tight for redhat bug #429689 to be fixed (5.2?).
-Jeff
J. Potter wrote:
I've been seeing the below message from yum whenever the repo has an update (CentOS 5):
/etc/cron.daily/yum.cron:
** Message: sqlite cache needs updating, reading in metadata
Googling a bit, it looks like others have seen this happen as well. The solutions, when I've found them, have been along the lines of "send output to /dev/null" or "edit this file", but nothing that feels like the "right" fix.
Is there a general solution to this? Or should we just sit tight for redhat bug #429689 to be fixed (5.2?).
this is not really a bug ...
it is just verbose output that causes an e-mail to be sent.
After 24 hours (the last time the script was run) the metadata is considered stale and too old to be used, so it is regenerated. By default, the program prints to standard out when it updates it's cache. So, you are getting that info in an e-mail as output of cron.
I do not see how this issue is at all related to RH bug 429689 ???
On Feb 19, 2008 8:14 AM, Johnny Hughes johnny@centos.org wrote:
I do not see how this issue is at all related to RH bug 429689 ???
Maybe it is 429869 ?
https://bugzilla.redhat.com/show_bug.cgi?id=429869
Akemi
** Message: sqlite cache needs updating, reading in metadata
... Is there a general solution to this? Or should we just sit tight for redhat bug #429689 to be fixed (5.2?).
this is not really a bug ... it is just verbose output that causes an e-mail to be sent.
It seems to be the default, though, and when managing 150+ servers, this ends up being a lot of email!
Is there a simple way to flip that message off?
I do not see how this issue is at all related to RH bug 429689 ???
Typo; meant 429869; as Akemi suggested.
-Jeff
J. Potter wrote:
** Message: sqlite cache needs updating, reading in metadata
... Is there a general solution to this? Or should we just sit tight for redhat bug #429689 to be fixed (5.2?).
how about doing a 'yum clean metadata; yum makecache' about 30 min before the yum.cron job is run ?
I do not see how this issue is at all related to RH bug 429689 ???
Typo; meant 429869; as Akemi suggested.
that bz also has absolutely nothing to do with your issue.