[CentOS-mirror] Big timeouts with unreachable mirrors.
anthony at somersettechsolutions.co.uk
Tue Sep 24 08:36:01 UTC 2013
mirror manager wont fix issues like that all the time
as that issue could be due to many reasons - some of which mirror manager may not pick up
for example routing issue between you and selected mirror
or if the "outage" occurs between mirror check probes
that error message below is your yum client timing out and moving on
you can tune this in /etc/yum.conf with the following variables
retries=N From the man pages: "6" is the default
timeout=N Where N is expressed in seconds
Somerset Technical Solutions Ltd.
Registered in the UK – Company no. 07738444
VAT Registration No: 140 6916 22
T: +44 (0) 33 0088 2751
E: anthony at somersettechsolutions.co.uk
On 24 Sep 2013, at 07:31, Martynas UAB Duomenų Centras <martynas at aleja.lt> wrote:
> Time to time some of the centos mirrors are unreachable (due many reasons).
> So on “yum ugrade” we got messages like:
> http://ftp.vectranet.pl/centos/6.4/updates/x86_64/repodata/fe8fa2beebe6c36885425555cf44699c00ffbc357c4323a7d0f01f86f0f8ac3c-primary.sqlite.bz2: [Errno 12] Timeout onhttp://ftp.vectranet.pl/centos/6.4/updates/x86_64/repodata/fe8fa2beebe6c36885425555cf44699c00ffbc357c4323a7d0f01f86f0f8ac3c-primary.sqlite.bz2: (28, 'Operation too slow. Less than 1 bytes/sec transfered the last 30 seconds')
> Trying other mirror.
> It takes a lot of time to wait while mirror check fails!
> Is in CentOS roadmap better mirror manager, to make users use nearest local mirrors, not from abroad!?
> Why yum fastest mirror plugin is not dropping server that don’t respond in three seconds, if it don’t respond for 30sec can it be fastest?
> CentOS-mirror mailing list
> CentOS-mirror at centos.org
-------------- next part --------------
An HTML attachment was scrubbed...
More information about the CentOS-mirror