;
> Another one that could make sense to monitor is "PoolFailed" in
> "show info". It should always remain zero.
We collect this (all available actually); I don't remember this one
ever measuring more than zero. But we'll keep an eye on it.
In time, could this be somewhat unrelated to HAProxy? I.e., maybe
kernel?
Cheers,
--
Ricardo Nabinger Sanchez https://www.taghos.com.br/
ecking if the default values are still
> in line with your usage.
We set our defaults pretty high in anticipation:
/proc/sys/fs/file-max = 5M;
/proc/sys/fs/nr_open = 3M;
Even with our software stack, we do not reach the limits. A long time
ago we did hit (lower limits ba
s/2508) with more details.
Cheers,
--
Ricardo Nabinger Sanchez https://www.taghos.com.br/
re trying to get a core dump, too.
Cheers,
--
Ricardo Nabinger Sanchez https://www.taghos.com.br/
5575d1f0) at
> src/checks.c:1549
This segfault might be fixed with this commit:
http://git.haproxy.org/?p=haproxy.git;a=commit;h=4bccea98912c74fa42c665ec25e417c2cca4eee7
Cheers,
--
Ricardo Nabinger Sanchez http://www.taghos.com.br/
the same behavior found in another function in
src/checks.c, which also calls protocol_by_family but that function
tests for a NULL pointer before dereferencing it.
Best regards,
--
Ricardo Nabinger Sanchez http://www.taghos.com.br/
>From 29684452b9cf923e321537ffc1f4694c88131f4b
d8b5b in __libc_start_main ()
from /lib64/libc.so.6 #5 0x0040257a in ?? ()
[Inferior 1 (process 4929) detached]
Note the use of "-batch", so there won't be GDB prompts such as
confirmation for detaching from the process.
Cheers,
--
Ricardo Nabinger Sanchez http://www.taghos.com.br/
t; Thus I believe this is a false-positive.
Thank you for checking this out and, by extension, Willy on his set of
replise. I missed the marker (and apparently, so did Clang).
Cheers,
--
Ricardo Nabinger Sanchez http://www.taghos.com.br/
to be correctly unearthed by
scan-build. Also, there might be a chance that this path invokes undefined
behavior, leading to further surprises.
Does this make sense?
Cheers,
--
Ricardo Nabinger Sanchez http://www.taghos.com.br/
ced by the filter 'cache'.\n",
925 curproxy->id, (char
*)fconf->conf);
926 err++;
927 }
I'm not sure how realistic this might be (especially to exploit), but it
feels worthy of reporting.
C
10 matches
Mail list logo