[Bug 66160] Apache HTTPD stops serving and freezes HTTP/2 connections

2022-07-11 Thread bugzilla 
https://bz.apache.org/bugzilla/show_bug.cgi?id=66160

--- Comment #1 from Stefan Eissing  ---
Hi Steffen,

the stacktrace you listed does not point to a deadlock. The conditional, timed
wait release the mutex. So it cannot be blocking others. If someone is holding
the mplx lock, it must be another thread.

Seeing h2 in this part of code is expected during connection processing. The
question is if your apache stops serving new connections or new requests on
existing connections. The former would point to an exhaustion of the mpm
workers, the latter to an exhaustion of the h2 worker pool.

Exhaustion can happen due to insufficient capacity or indeed a bug in the
server that blocks processing and holds resources too long.

The most tricky things in a HTTP server are long running requests. If your
setup/backend possibly encounters those, they can clog up your system. HTTP/2
is more vulnerable to that, if your h2 worker pool is "too small".

Hope this helps,

Stefan

-- 
You are receiving this mail because:
You are the assignee for the bug.
-
To unsubscribe, e-mail: bugs-unsubscr...@httpd.apache.org
For additional commands, e-mail: bugs-h...@httpd.apache.org

[Bug 66168] New: Segmentation fault

2022-07-11 Thread bugzilla 
https://bz.apache.org/bugzilla/show_bug.cgi?id=66168

Bug ID: 66168
   Summary: Segmentation fault
   Product: Apache httpd-2
   Version: 2.4.53
  Hardware: PC
OS: Linux
Status: NEW
  Severity: normal
  Priority: P2
 Component: mod_brotli
  Assignee: bugs@httpd.apache.org
  Reporter: apa...@mfai.li
  Target Milestone: ---

Created attachment 38339
  --> https://bz.apache.org/bugzilla/attachment.cgi?id=38339=edit
steps to reproduce

We've stumbled upon segmentation faults on some of our ubuntu servers
(18.04/20.04/22.04). It seems to only occur when the DocumentRoot is outside of
/var/www/html.
This segmentation fault leads to killing the parent server process with its
childs, resulting in clients getting "Empty reply".

We were able to replicate the issue on Apache versions 2.4.41, 2.4.52 and
2.4.54 (mpm_event).
Our Apache is configured to work with php-fpm over proxy_fcgid. We could
replicate the issue with php versions 5.6-7.4, where the script-output is a lot
of: "Warning: Use of undefined constant...".

I've attached the steps to reproduce with a "fresh" ubuntu installation.

 Errorlog 

apache2[2470393]: [core:notice] [pid 2470393:tid 140639993146432] AH00051:
child pid 2473145 exit signal Segmentation fault (11), possible coredump in
/var/tmp

 CoreDump 

Core was generated by `/usr/sbin/apache2 -k start'.
Program terminated with signal SIGSEGV, Segmentation fault.
#0  0x7f1b5a487056 in BrotliEncoderCompressStream () from
/lib/x86_64-linux-gnu/libbrotlienc.so.1
[Current thread is 1 (Thread 0x7f1b46328700 (LWP 2404021))]
(gdb) bt
#0  0x7f1b5a487056 in BrotliEncoderCompressStream () from
/lib/x86_64-linux-gnu/libbrotlienc.so.1
#1  0x7f1b5a53bac0 in process_chunk (f=0x7f1b4028d900, f=0x7f1b4028d900,
len=184, data=, ctx=0x7f1b4028eed0) at mod_brotli.c:222
#2  compress_filter (f=0x7f1b4028d900, bb=0x7f1b4028eaf8) at mod_brotli.c:561
#3  0x7f1b5a50d2d0 in filter_harness (f=, bb=0x7f1b4028eaf8)
at mod_filter.c:323
#4  0x7f1b5a315e6e in dispatch (conn=conn@entry=0x7f1b566060a0,
conf=conf@entry=0x7f1b56b01ad8, r=r@entry=0x7f1b402b60a0,
setaside_pool=0x7f1b40289028, err=err@entry=0x7f1b46327b40,
bad_request=bad_request@entry=0x7f1b46327b08, 
has_responded=0x7f1b46327b0c, input_brigade=0x7f1b4028dbd8, request_id=1)
at mod_proxy_fcgi.c:849
#5  0x7f1b5a316e7c in fcgi_do_request (p=, origin=0x0,
uri=, url=, input_brigade=0x7f1b4028dbd8,
server_portstr=0x7f1b46327bb0 "", conf=0x7f1b56b01ad8, conn=0x7f1b566060a0, 
r=0x7f1b402b60a0) at mod_proxy_fcgi.c:992
#6  proxy_fcgi_handler (r=0x7f1b402b60a0, worker=,
conf=, url=, proxyname=,
proxyport=) at mod_proxy_fcgi.c:1206
#7  0x7f1b5a2da31b in proxy_run_scheme_handler (r=r@entry=0x7f1b402b60a0,
worker=0x7f1b56da2358, conf=conf@entry=0x7f1b56da2018, url=0x7f1b4028dad0
"fcgi://php74fpm/www/htdocs/test3.php", 
proxyhost=proxyhost@entry=0x0, proxyport=proxyport@entry=0) at
mod_proxy.c:3434
#8  0x7f1b5a2dc3cb in proxy_handler (r=0x7f1b402b60a0) at mod_proxy.c:1507
#9  0x559dc8bc7268 in ap_run_handler (r=r@entry=0x7f1b402b60a0) at
config.c:169
#10 0x559dc8bc7816 in ap_invoke_handler (r=r@entry=0x7f1b402b60a0) at
config.c:443
#11 0x559dc8be0053 in ap_process_async_request (r=0x7f1b402b60a0) at
http_request.c:452
#12 0x559dc8be0222 in ap_process_request (r=r@entry=0x7f1b402b60a0) at
http_request.c:487
#13 0x7f1b5a3a6f69 in h2_task_process_request (c=0x7f1b402bc0a0,
task=) at h2_task.c:671
#14 h2_task_process_conn (c=0x7f1b402bc0a0) at h2_task.c:713
#15 h2_task_process_conn (c=0x7f1b402bc0a0) at h2_task.c:700
#16 0x559dc8bd1008 in ap_run_process_connection (c=c@entry=0x7f1b402bc0a0)
at connection.c:42
#17 0x7f1b5a3a82c7 in h2_task_do (task=0x7f1b402b80a0,
thread=thread@entry=0x7f1b59b40090, worker_id=) at h2_task.c:631
#18 0x7f1b5a3abf80 in slot_run (thread=0x7f1b59b40090, wctx=0x7f1b56dcf420)
at h2_workers.c:263
#19 0x7f1b5b2b6609 in start_thread () from
/lib/x86_64-linux-gnu/libpthread.so.0
#20 0x7f1b5b1db133 in clone () from /lib/x86_64-linux-gnu/libc.so.6

-- 
You are receiving this mail because:
You are the assignee for the bug.
-
To unsubscribe, e-mail: bugs-unsubscr...@httpd.apache.org
For additional commands, e-mail: bugs-h...@httpd.apache.org