> > Hi,
> >
> > Since the switch from libsetrans to mcstrans (0.1.1-1 and 0.1.2-1 tested)
> two
> > days ago, I'm experiencing several issues on my Itanium workstation. During
> > startup, sm-client takes a long time (1-2 minutes) to start. Well in fact,
> it
> > fails to start with the following error:
> >
> > file_contexts: invalid context system_u:object_r:sendmail_var_run_t:s0
> > matchpathcon (/var/run/sm-client.pid) failed Invalid argument
> >
> > More annoying, xfs fails to start too and thus gdm can't start the X
> server.
> > I've also noticed that the HDD activity LED keeps busy. Indeed, my
> > /var/log/messages file is flooded with lines like these:
> >
> > May 13 00:22:05 localhost mcstransd: Could not allocate 1953724787 bytes
> > May 13 00:22:05 localhost mcstransd: Servicing of request failed for fd (4)
> > May 13 00:22:05 localhost mcstransd: Could not allocate 1953724787 bytes
> > May 13 00:22:05 localhost mcstransd: Servicing of request failed for fd (6)
> >
> > Always the same lines with fd (4) and fd (6).
> >
> > Due to this abnormal HDD activity, my system is not very responsive. Once
> > mcstransd is manually stopped, I can restart xfs and then the X server.
> >
> > The Fedora Devel list or bugzilla have no information regarding these
> issues.
> > Are they specific to the ia64 platform?
>
> I don't think so, I'm experiencing same issue on x86_64. Hopefully
> tomorrow's update will cure it.
Problem still present on ia64 with today's updates: mcstrans-0.1.3-1,
selinux-policy-*-2.2.38-6 and autit-libs-*-1.2.2-1.
Émeric
--
fedora-devel-list mailing list
fedora-devel-list@xxxxxxxxxx
https://www.redhat.com/mailman/listinfo/fedora-devel-list
