Bug#914606: apache2 setup-instance (apache-multi) logrotation
Package: apache2
Version: 2.4.25-3+deb9u6
Severity: normal
Dear Maintainer,
there is a problem with the setup-instance script for apache-multi
installation. The logrotation for the copyied instance is not working.
It looks like it is in debian stretch and buster/sid, in a kind of the
same way. In my fokus it's comming up with an installation from Ubuntu 18.04
What i did (stretch; buster/sid; ubuntu 18.04)
:~# apt-get install apache2
[...]
:~# sh /usr/share/doc/apache2/examples/setup-instance multi-xxx
Setting up /etc/apache2-multi-xxx ...
systemd is in use, no init script installed
use the 'apache2@multi-xxx.service' service to control your new instance
sample commands:
systemctl start apache2@multi-xxx.service
systemctl enable apache2@multi-xxx.service
Setting up symlinks: a2enmod-multi-xxx a2dismod-multi-xxx
a2ensite-multi-xxx a2dissite-multi-xxx a2enconf-multi-xxx
a2disconf-multi-xxx apache2ctl-multi-xxx
HINT "no init script installed"
:~# ls -la /etc/init.d/apache*
-rwxr-xr-x 1 root root 8181 Nov 3 12:34 /etc/init.d/apache2
-rwxr-xr-x 1 root root 2489 Nov 3 12:34 /etc/init.d/apache-htcacheclean
-> No init.d for the new multi-xxx apache
With in Debian 9.6 stretch
:~# cat /etc/logrotate.d/apache2-multi-xxx
/var/log/apache2-multi-xxx/*.log {
daily
missingok
rotate 14
compress
delaycompress
notifempty
create 640 root adm
sharedscripts
postrotate
if /etc/init.d/apache2-multi-xxx status > /dev/null ; then \
/etc/init.d/apache2-multi-xxx reload > /dev/null; \
fi;
endscript
prerotate
if [ -d /etc/logrotate.d/httpd-prerotate ]; then \
run-parts /etc/logrotate.d/httpd-prerotate; \
fi; \
endscript
}
:~# /etc/init.d/apache2-multi-xxx status
-su: /etc/init.d/apache2-multi-xxx: Datei oder Verzeichnis nicht gefunden
:~# /etc/init.d/apache2-multi-xxx reload
-su: /etc/init.d/apache2-multi-xxx: Datei oder Verzeichnis nicht gefunden
With in Debian buster/sid
:~$ cat /etc/logrotate.d/apache2-multi-xxx
/var/log/apache2-multi-xxx/*.log {
daily
missingok
rotate 14
compress
delaycompress
notifempty
create 640 root adm
sharedscripts
postrotate
if invoke-rc.d apache2-multi-xxx status > /dev/null
2>&1; then \
invoke-rc.d apache2-multi-xxx reload > /dev/null 2>&1; \
fi;
endscript
prerotate
if [ -d /etc/logrotate.d/httpd-prerotate ]; then \
run-parts /etc/logrotate.d/httpd-prerotate; \
fi; \
endscript
}
:~# invoke-rc.d apache2-multi-xxx status
Unit apache2-multi-xxx.service could not be found.
invoke-rc.d: initscript apache2-multi-xxx, action "status" failed.
:~# invoke-rc.d apache2-multi-xxx reload
/usr/sbin/invoke-rc.d: 527: /usr/sbin/invoke-rc.d:
/etc/init.d/apache2-multi-xxx: not found
invoke-rc.d: initscript apache2-multi-xxx, action "reload" failed.
In both cases the logrotation will fail because the reload is not
working. Also in the both cases the logrotate.d script for the multi-xxx
apache is a copy from the original logrotate.d apache2 script.
:~$ cat /usr/share/doc/apache2/examples/setup-instance
[...]
echo Setting up /etc/logrotate.d/apache2-$SUFFIX and
/var/log/apache2-$SUFFIX ...
cp -a /etc/logrotate.d/apache2 /etc/logrotate.d/apache2-$SUFFIX
perl -p -i -e s,apache2,apache2-$SUFFIX,g /etc/logrotate.d/apache2-$SUFFIX
[...]
As shown above the original apache has a init.d script installed and the
logrotation is working.
In that Ubuntu 18.04 installation i changed the
/etc/logrotate.d/apache2-multi-xxx in that way that i changed the reload
to systemd
from
[...]
if invoke-rc.d apache2-multi-xxx status > /dev/null
2>&1; then \
invoke-rc.d apache2-multi-xxx reload > /dev/null 2>&1; \
[...]
in to
[...]
if systemctl status apache2@multi-xxx.service >
/dev/null 2>&1; then \
systemctl reload apache2@multi-xxx.service >
/dev/null 2>&1; \
[...]
Looks like some day's after the logrotation is now workings as expected
for the multi-instance.
And in that case i think the upstream package for ubuntu 18.04 is
comming from Debian buster/sid.
Kind regards
Horst
Processed: severity of 914297 is serious
Processing commands for cont...@bugs.debian.org: > severity 914297 serious Bug #914297 [apache2] apache2: getrandom call blocks on first startup, systemd kills with timeout Severity set to 'serious' from 'important' > thanks Stopping processing here. Please contact me if you need assistance. -- 914297: https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=914297 Debian Bug Tracking System Contact ow...@bugs.debian.org with problems
Bug#840580: apache2-bin: crashes when issuing a restart while mod_cgid is enabled
Dear Maintainer, tried to find out the actual location that the backtrace points to. Unfortunately I could not make any clue out of the line containing /usr/sbin/apache2(+0x29e450). But at least, I think, the line containing mod_mpm_prefork.so(+0x4c08) translates to function prefork_run in server/mpm/prefork/prefork.c. As this is a rather big function, and looks like it is never left while the server runs, and there are no local arrays accessed, the stack canary may be overwritten by some function called from there. But the stack canary is just checked when prefork_run exits. Kind regards, Bernhard *** stack smashing detected ***: /usr/sbin/apache2 terminated === Backtrace: = /lib/x86_64-linux-gnu/libc.so.6(+0x731af)[0x7f6d8e1c11af]| 0x7f6d8e1c11af | /lib/x86_64-linux-gnu/libc.so.6(__fortify_fail+0x37)[0x7f6d8e246aa7] | 0x7f6d8e246aa7 | /lib/x86_64-linux-gnu/libc.so.6(__fortify_fail+0x0)[0x7f6d8e246a70] | 0x7f6d8e246a70 | /usr/lib/apache2/modules/mod_mpm_prefork.so(+0x4c08)[0x7f6d8b462c08] | 0x7f6d8b462c08 | 0x7f6193a75c08: 0x7f6193a75c03 : callq 0x7f6193a73400 <__stack_chk_fail@plt> /usr/sbin/apache2(+0x29e450)[0x7f6d8f2a3450] | 0x7f6d8f2a3450 | === Memory map: 7f6d8f005000-7f6d8f09d000 r-xp fe:00 3882 /usr/sbin/apache2 7f6d8b45e000-7f6d8b465000 r-xp fe:00 127839 /usr/lib/apache2/modules/mod_mpm_prefork.so apt install dpkg-dev devscripts mc gdb binutils apache2-bin apache2-dbg # http://snapshot.debian.org/package/apache2/2.4.10-10%2Bdeb8u7/ wget http://snapshot.debian.org/archive/debian/20160916T101556Z/pool/main/a/apache2/apache2-bin_2.4.10-10%2Bdeb8u7_amd64.deb wget http://snapshot.debian.org/archive/debian/20160916T101556Z/pool/main/a/apache2/apache2-dbg_2.4.10-10%2Bdeb8u7_amd64.deb dpkg -i --force-depends apache2-bin_2.4.10-10+deb8u7_amd64.deb apache2-dbg_2.4.10-10+deb8u7_amd64.deb mkdir apache2/orig -p cdapache2/orig dget http://snapshot.debian.org/archive/debian/20160916T101556Z/pool/main/a/apache2/apache2_2.4.10-10%2Bdeb8u7.dsc dpkg-source -x apache2_2.4.10-10%2Bdeb8u7.dsc cd ../.. a2dismod mpm_event a2enmod mpm_prefork systemctl stop apache2 systemctl start apache2 root@debian:~# gdb -q --pid 16415 ... (gdb) set width 0 (gdb) set pagination off (gdb) directory /home/benutzer/apache2/orig/apache2-2.4.10/server Source directories searched: /home/benutzer/apache2/orig/apache2-2.4.10/server:$cdir:$cwd (gdb) b main Breakpoint 1 at 0x556c539ec940: file main.c, line 439. (gdb) disassemble prefork_run,prefork_run+3830 Dump of assembler code from 0x7f6193a74d60 to 0x7f6193a75c56: 0x7f6193a74d60 : push %r15 ... 0x7f6193a74d81 : mov%fs:0x28,%rax ; Value loaded into $rax 0x7f6193a74d8a : mov%rax,0xe8(%rsp) ; Value stored in canary ... 0x7f6193a75288 : mov0xe8(%rsp),%rbx ; Canary loaded into $rbx 0x7f6193a75290 : xor%fs:0x28,%rbx ; Canary compared to the original value 0x7f6193a75299 : mov%r13d,%eax 0x7f6193a7529c : jne0x7f6193a75c03 ... 0x7f6193a75c03 : callq 0x7f6193a73400 <__stack_chk_fail@plt> 0x7f6193a75c08 : callq 0x7f6193a73300 <__errno_location@plt> ... 0x7f6193a75c4b : jmpq 0x7f6193a75b9c 0x7f6193a75c50 : push %rbp End of assembler dump. set width 0 set pagination off directory /home/benutzer/apache2/orig/apache2-2.4.10/server b main run
Bug#840580: apache2-bin: crashes when issuing a restart while mod_cgid is enabled
On Sunday, 25 November 2018 19:07:56 CET Bernhard Übelacker wrote: > Dear Maintainer, > tried to find out the actual location that the backtrace points to. > > Unfortunately I could not make any clue out of the line > containing /usr/sbin/apache2(+0x29e450). > > But at least, I think, the line containing mod_mpm_prefork.so(+0x4c08) > translates to function prefork_run in server/mpm/prefork/prefork.c. > > As this is a rather big function, and looks like it is never left while > the server runs, and there are no local arrays accessed, the stack > canary may be overwritten by some function called from there. > But the stack canary is just checked when prefork_run exits. > > Kind regards, > Bernhard The "bt full" gdb command is the important one. Please show its output as described in README.backtrace
Bug#914297: apache2: getrandom call blocks on first startup, systemd kills with timeout
How long is the timeout after which it is killed? What is the status of systemd-random-seed.service in that case? I don't see why it should take so long for the random number generator to initialize. But maybe apache2 needs to add a dependency. Please provide the output of journalctl -b UNIT=apache2.service + UNIT=systemd-random-seed.service + _TRANSPORT=kernel|grep -i -e apache -e random when apache2 has failed to start.
