[Yum] YUM timing out trying to get updates from Fedora mirrors

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



On Tue, Feb 17, 2004 at 04:24:59PM -0500, Jim McConnell wrote:
> on 2/17/2004 6:23 AM J?rgen M?llenhoff said the following:
> >Is it possible to create some kind of timeout for the download, so that 
> >yum gives up when there is no response for an hour or so? In the moment 
> >yum sits there forerver with no response at all, in such a situation I 
> >must always kill yum manually.

Yes.  It is possible, and it is in development.

> I've seen this as well.  I have several mirrors listed in yum.conf. 
> Since I'm still testing yum, I've been running it interactively.  When 
> this happens to me, I'm able to ^C, and the download restarts presumably 
> from another mirror.  I haven't checked this via netstat, but I think 
> that is what is happening.  Anyway, I don't know if this is a feature or 
> not, YMMV.

Not an intentional feature, but handy to know.

> Could a developer comment on this behavior?  Is it normal?

It is not intentional and it may be doing something evil underneath,
but I doubt it.

> Also, as a "feature request", is it possible in python to do something 
> similar to a "trap"?  When I hit ^C, I often get a dump of what line in 
> the code I was at, etc.  Not a happy thing for most people to see, I'd 
> imagine.  Dumping that in the log might be useful...

It is possible to "trap" it.  In fact, that is likely what's happening
in the above case.  It's sitting there waiting for some network event,
you send it SIGINT, python raises an exception, the exception is
caught and probably mistaken for a network error, and so the next
mirror (or retry) as attempted.

Yum catches KeyboardInterrupts in most places.  Often you will see
"Interrupted by user" or something.  If you instead see a traceback,
then that should probably be considered a bug, and you can feel free
to submit it to bugzilla.  Be sure to include the actual traceback so
we can see where it happened.

It is my humble opinion (and it may not be Seth's) that any time a
user sees a traceback on stderr, it's a bug.  That said, it's often
most efficient to "catch them as they come".  That is, don't try and
anticipate all the crazy things that can happen.  Deal with them when
they occur.

					-Michael
-- 
  Michael D. Stenner                            mstenner@xxxxxxxxxxxxxxx
  ECE Department, the University of Arizona                 520-626-1619
  1230 E. Speedway Blvd., Tucson, AZ 85721-0104                 ECE 524G

[Index of Archives]     [Fedora Users]     [Fedora Legacy List]     [Fedora Maintainers]     [Fedora Desktop]     [Fedora SELinux]     [Big List of Linux Books]     [Yosemite News]     [KDE Users]

  Powered by Linux