Re: [Savannah-hackers-public] git accumulations on vcs

[James Cloos]

>>>>> "KB" == Karl Berry <address@hidden> writes:

KB> Ha!  Brilliant sleuthing.

Once I read the timeout(1) man page and info chapter it was obvious that
the git daemon might be ignoring the signal; git-grep(1) confirmed that.

KB> I never knew that about SIGRTMIN.  Will remember the next time the
KB> dreaded D wait arises.

Any of the realtime signals, from SIGRTMIN to SIGRTMAX usually will
break out of a D state; when typing at the shell I usually use -55
because it is easy to remember and type.  -44 would be as easy.
(On linux amd64 -- this is a kernel thing not userland -- the realtime
signals are 34 to 64, inclusive.

KB> Latest hang was for 58.247.233.24, by the way.  The last named host
KB> in traceroute is cnc-gw.customer.alter.net.  Hmm.

whois 58.247.233.24 reports CHINA UNICOM Shanghai network.

I'll write a note to the git list about the GFW's affect on sv's gitd.

-JimC
-- 
James Cloos <address@hidden>         OpenPGP: 1024D/ED7DAEA6