enlistments stuck after temp. failure

Hi,

the eurynome.mirbsd.org enlistments, such as mirbsd, seem to be stuck. We had a misconfiguration in the cvs pee-server for a while a few weeks or so ago, but that has been fixed long since. The downloader seems to not have recognised that. (This applies to almost all projects with an en- listment on eurynome.)

I'd suggest exponential retries (1, 2, 4, 8 days, biweekly after that) on server failures (misconfiguration, EUNREACH due to system failure or maintenance, etc.) to aid against such problems in the future. For example, eurynome is a VM whose host is rebooted every once in a while (say twice in 3 months) but that's out of my control, so we'll obviously have a little downtime.

Still broken, e.g. one of the mksh enlistments too. https://www.ohloh.net/p/mirbsd/enlistments I already mentioned.

Thanks, I saw the cvs processes running, and they completed. Now, the two mirports enlistments and one of the mksh ones, and maybe others (basically all on eurynome, I suppose).

The cvs process does appear to be very slow, judging from the server side view. I wonder what you’re actually doing?

My offer of rsync access to the repo so that you can do that kind of things locally still stands (although I’d have to break up some of the symlink farms I created, for example with mksh, for that).

MirPorts and one of the mksh ones still fail.

Yes, that was my configuration problem I had already fixed quite some time ago.

Now there are still some failing repos, for example one of the mksh ones. The MirPorts repo ran through.