[Maria-discuss] Bad file descriptor mariadb 10.1.
Hi List, Sometimes, my mysqld daemon begin to consumes 100% the connection to the server gets unstable and I get flooded in the mysql error log with these messages: Nov 4 06:17:01 db1 mysqld: 2015-11-04 6:17:01 140682979031168 [ERROR] Error in accept: Bad file descriptor Nov 4 06:17:01 db1 mysqld: 2015-11-04 6:17:01 140682979031168 [ERROR] Error in accept: Bad file descriptor Nov 4 06:17:01 db1 mysqld: 2015-11-04 6:17:01 140682979031168 [ERROR] Error in accept: Bad file descriptor Nov 4 06:17:01 db1 mysqld: 2015-11-04 6:17:01 140682979031168 [ERROR] Error in accept: Bad file descriptor Nov 4 06:17:01 db1 mysqld: 2015-11-04 6:17:01 140682979031168 [ERROR] Error in accept: Bad file descriptor Nov 4 06:17:01 db1 mysqld: 2015-11-04 6:17:01 140682979031168 [ERROR] Error in accept: Bad file descriptor it only stops when I restart the daemon, no query is shown in "show full processlist". Maybe some kernel adjustment? Any ideia? Running centos 7 at the lastest version. [root@db1 scripts]# yum -y update Loaded plugins: fastestmirror Loading mirror speeds from cached hostfile * base: ftpmirror.your.org * epel: mirror.nexcess.net * extras: ftpmirror.your.org * updates: mirrors.lga7.us.voxel.net No packages marked for update [root@db1 scripts]# rpm -qa|grep -i MariaDB MariaDB-shared-10.1.8-1.el7.centos.x86_64 MariaDB-client-10.1.8-1.el7.centos.x86_64 MariaDB-common-10.1.8-1.el7.centos.x86_64 MariaDB-server-10.1.8-1.el7.centos.x86_64 Thanks! João Reis.
On 8 Nov 2015, at 03:36, joao@7lan.net wrote:
Sometimes, my mysqld daemon begin to consumes 100% the connection to the server gets unstable and I get flooded in the mysql error log with these messages:
Nov 4 06:17:01 db1 mysqld: 2015-11-04 6:17:01 140682979031168 [ERROR] Error in accept: Bad file descriptor
Sometimes, is a problem ;-) Its much better if we can always repeat it what is open_files_limit set to in my.cnf? also what is the ulimit set at? -- Colin Charles, Chief Evangelist, MariaDB Corporation blog: http://bytebot.net/blog/ | t: +1-347-903-3201 | Skype: colincharles | Twitter: @bytebot
Yep, without a way to reproduce the error it gets funnier :D BUT, I remembered something, this is a new installation of an old server and I got some similar errors in the past. I saved the old configuration and I have this: [mysqld] open_files_limit = 24000 This new server have the default open_files_limit config. MariaDB [(none)]> show global variables like 'open%'; +------------------+-------+ | Variable_name | Value | +------------------+-------+ | open_files_limit | 1024 | +------------------+-------+ 1 row in set (0.01 sec) MariaDB [(none)]> Maybe thats the problem... I think because the many files tokudb uses in my datadir. will increase and see what happens.. If you guys have any additional information that might help please let me know. [root@db1 ~]# ulimit -a core file size (blocks, -c) 0 data seg size (kbytes, -d) unlimited scheduling priority (-e) 0 file size (blocks, -f) unlimited pending signals (-i) 15057 max locked memory (kbytes, -l) 64 max memory size (kbytes, -m) unlimited open files (-n) 1024 pipe size (512 bytes, -p) 8 POSIX message queues (bytes, -q) 819200 real-time priority (-r) 0 stack size (kbytes, -s) 8192 cpu time (seconds, -t) unlimited max user processes (-u) 15057 virtual memory (kbytes, -v) unlimited file locks (-x) unlimited Thanks! 7 de novembro de 2015 às 18:05, "Colin Charles" <colin@mariadb.org> escreveu:
On 8 Nov 2015, at 03:36, joao@7lan.net wrote:
Sometimes, my mysqld daemon begin to consumes 100% the connection to the server gets unstable and I get flooded in the mysql error log with these messages:
Nov 4 06:17:01 db1 mysqld: 2015-11-04 6:17:01 140682979031168 [ERROR] Error in accept: Bad file descriptor
Sometimes, is a problem ;-) Its much better if we can always repeat it
what is open_files_limit set to in my.cnf? also what is the ulimit set at? -- Colin Charles, Chief Evangelist, MariaDB Corporation blog: http://bytebot.net/blog | t: +1-347-903-3201 | Skype: colincharles | Twitter: @bytebot
Am 07.11.2015 um 21:19 schrieb joao@7lan.net:
I saved the old configuration and I have this:
[mysqld] open_files_limit = 24000
This new server have the default open_files_limit config.
MariaDB [(none)]> show global variables like 'open%'; +------------------+-------+ | Variable_name | Value | +------------------+-------+ | open_files_limit | 1024 | +------------------+-------+ 1 row in set (0.01 sec)
why in the world do you start with a default config and after that wonder when something breaks? the last time i wrote a my.cnf from scratch was 2003
On 8 Nov 2015, at 04:19, joao@7lan.net wrote:
Yep, without a way to reproduce the error it gets funnier :D
BUT, I remembered something, this is a new installation of an old server and I got some similar errors in the past.
I saved the old configuration and I have this:
[mysqld]
open_files_limit = 24000
This new server have the default open_files_limit config.
MariaDB [(none)]> show global variables like 'open%'; +------------------+-------+ | Variable_name | Value | +------------------+-------+ | open_files_limit | 1024 | +------------------+-------+ 1 row in set (0.01 sec)
MariaDB [(none)]>
Maybe thats the problem... I think because the many files tokudb uses in my datadir.
will increase and see what happens.. If you guys have any additional information that might help please let me know.
also worth noting: https://mariadb.atlassian.net/browse/MDEV-8995 -- Colin Charles, Chief Evangelist, MariaDB Corporation blog: http://bytebot.net/blog/ | t: +1-347-903-3201 | Skype: colincharles | Twitter: @bytebot
Most likely it is that ulimit is too low. Try the instructions in this blog post (not mine): https://ma.ttias.be/increase-open-files-limit-in-mariadb-on-centos-7-with-sy... Sent from my iPhone
On Nov 7, 2015, at 12:36 PM, joao@7lan.net wrote:
Hi List,
Sometimes, my mysqld daemon begin to consumes 100% the connection to the server gets unstable and I get flooded in the mysql error log with these messages:
Nov 4 06:17:01 db1 mysqld: 2015-11-04 6:17:01 140682979031168 [ERROR] Error in accept: Bad file descriptor Nov 4 06:17:01 db1 mysqld: 2015-11-04 6:17:01 140682979031168 [ERROR] Error in accept: Bad file descriptor Nov 4 06:17:01 db1 mysqld: 2015-11-04 6:17:01 140682979031168 [ERROR] Error in accept: Bad file descriptor Nov 4 06:17:01 db1 mysqld: 2015-11-04 6:17:01 140682979031168 [ERROR] Error in accept: Bad file descriptor Nov 4 06:17:01 db1 mysqld: 2015-11-04 6:17:01 140682979031168 [ERROR] Error in accept: Bad file descriptor Nov 4 06:17:01 db1 mysqld: 2015-11-04 6:17:01 140682979031168 [ERROR] Error in accept: Bad file descriptor
it only stops when I restart the daemon, no query is shown in "show full processlist".
Maybe some kernel adjustment? Any ideia?
Running centos 7 at the lastest version.
[root@db1 scripts]# yum -y update Loaded plugins: fastestmirror Loading mirror speeds from cached hostfile * base: ftpmirror.your.org * epel: mirror.nexcess.net * extras: ftpmirror.your.org * updates: mirrors.lga7.us.voxel.net No packages marked for update [root@db1 scripts]# rpm -qa|grep -i MariaDB MariaDB-shared-10.1.8-1.el7.centos.x86_64 MariaDB-client-10.1.8-1.el7.centos.x86_64 MariaDB-common-10.1.8-1.el7.centos.x86_64 MariaDB-server-10.1.8-1.el7.centos.x86_64
Thanks!
João Reis.
_______________________________________________ Mailing list: https://launchpad.net/~maria-discuss Post to : maria-discuss@lists.launchpad.net Unsubscribe : https://launchpad.net/~maria-discuss More help : https://help.launchpad.net/ListHelp
participants (4)
-
Colin Charles
-
joao@7lan.net
-
Justin Swanhart
-
Reindl Harald