[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: CVS problem with ssh
From: |
Alexander Taler |
Subject: |
Re: CVS problem with ssh |
Date: |
Tue, 19 Jul 2005 23:26:03 -0400 |
>>>>> "Paul" == Paul Eggert <eggert@CS.UCLA.EDU> writes:
Paul> To: Derek Price <derek@ximbiot.com>
Paul> Subject: Re: CVS problem with ssh
Paul> Date: Sat, 16 Jul 2005 18:13:03 -0700
Paul> OK, here's a proposed patch to CVS 1.12.12 for the CVS + ssh problem.
Paul> It uses a pipe, but leaves stdio alone.
Hello,
I intermittently receive a failure in the test sshstdio-6 from sanity.sh on my
OpenBSD 3.6 sparc64 machine. Am I correct in believing that this is the bug
under discussion?
In any case, I tried your patch Paul, but it did not correct the bug, instead
CVS just hangs. I obtained stack traces from both the parent and child
processes, in a somewhat unorthodox way. gdb would hang when I attempted to
attach it to the processes, so instead I sent SIGBUS to generate core files
(SIGABRT didn't work). I am somewhat mystified by the stack traces, which is
why I haven't tried further debugging, but I am no system programmer.
I applied the patch to the HEAD of CVS, not to the 1.12.12 release itself, so
the line numbers may be slightly off.
Here's the stack trace from the parent:
(gdb) where
#0 0x0000000040c5b4c8 in vfork () from /usr/lib/libc.so.34.1
#1 0x0000000000156720 in piped_child (command=0xffffffffffffa010,
tofdp=0xffffffffffffa00c, fromfdp=0xffffffffffffa008, fix_stderr=true) at
run.c:558
#2 0x0000000000155638 in start_rsh_server (root=0x7d2000,
to_server_p=0x7c49c8, from_server_p=0x7c49d0) at rsh-client.c:187
#3 0x0000000000113238 in open_connection_to_server (root=0x7d2000,
to_server_p=0x7c49c8, from_server_p=0x7c49d0) at client.c:3906
#4 0x00000000001132ec in start_server () at client.c:3967
#5 0x000000000011aa44 in diff (argc=1, argv=0x7c8248) at diff.c:410
#6 0x0000000000139a04 in main (argc=10, argv=0x7c8200) at main.c:1066
And from the child:
(gdb) where
#0 0x00000000435ddc28 in read () from /usr/lib/libc.so.34.1
#1 0x00000000001563ac in work_around_openssh_glitch () at run.c:482
#2 0x0000000000156874 in piped_child (command=0xffffffffffffaac0,
tofdp=0xffffffffffffaabc, fromfdp=0xffffffffffffaab8, fix_stderr=true) at
run.c:576
#3 0x0000000000155638 in start_rsh_server (root=0x7d2000,
to_server_p=0x7c49c8, from_server_p=0x7c49d0) at rsh-client.c:187
#4 0x0000000000113238 in open_connection_to_server (root=0x7d2000,
to_server_p=0x7c49c8, from_server_p=0x7c49d0) at client.c:3906
#5 0x00000000001132ec in start_server () at client.c:3967
#6 0x000000000011aa44 in diff (argc=1, argv=0x7c8248) at diff.c:410
#7 0x0000000000139a04 in main (argc=10, argv=0x7c8200) at main.c:1066
Also, I may not be responding to email for the next week, but I will certainly
do what I can when I can get to the network.
Alex
--
https://savannah.gnu.org/projects/libcvs-spec Access CVS through a library.
PGP: ID: 0x23DC453B FPR: 42D0 66C2 9FF8 553A 373A B819 4C34 93BA 23DC 453B
The pimp's trade must be carried out by intelligent people, is essential to any
well-ordered society, and should have an official inspector. -- Don Quixote
- Re: CVS problem with ssh, (continued)
- Re: CVS problem with ssh, Paul Eggert, 2005/07/13
- Re: CVS problem with ssh, Richard M. Stallman, 2005/07/14
- Re: CVS problem with ssh, Paul Eggert, 2005/07/14
- Re: CVS problem with ssh, Richard M. Stallman, 2005/07/15
- Re: CVS problem with ssh, Larry Jones, 2005/07/15
- Re: CVS problem with ssh, Richard M. Stallman, 2005/07/15
- Re: CVS problem with ssh, Larry Jones, 2005/07/15
- Re: CVS problem with ssh, Richard M. Stallman, 2005/07/16
- Re: CVS problem with ssh, Derek Price, 2005/07/15
- Message not available
- Message not available
- Re: CVS problem with ssh, Paul Eggert, 2005/07/16
- Re: CVS problem with ssh,
Alexander Taler <=
Re: CVS problem with ssh, Richard M. Stallman, 2005/07/13
Re: CVS problem with ssh, Paul Eggert, 2005/07/16