1H Troubleshooting

Steps to resolve 1H issues:

1. Ensure that it is a hive problem.
2. Follow frequently seen cases.
3. Forward to operations.

About point 1. If it is possible try to disable the chroot and limits for this account using apache mod_hive options (
http:docs.1h.com/Mod_hive:_options ) or crond excludes (
http:docs.1h.com/Crond#Exclude_an_account_from_global_limits ) if the problem persists even with hive disabled for
this account, most probably there is another issue.

About point 2.
If the issue is 'Unable to connect to MySQL':
1. Check if MySQL is running on the machine.
2. Check if MySQL has /tmp/mysql.sock open (this is done either with lsof +p pid_of_mysql or ls -l
/proc/PID_OF_MYSQL/fs/ )
3. Check if MySQL is configured to listen to /tmp/mysql.sock (mysql -e "show variables like 'socket'" and also grep
sock /etc/my.cnf )
If the issues is 'ISE not related to MySQL':

1. Check if all mounts for the specific user are there:
# grep ayannka /proc/mounts
none /var/suexec/ayannka tmpfs rw,relatime,size=614400k 0 0
/dev/sda4 /var/suexec/ayannka/chroot ext4 rw,relatime,barrier=1,data=ordered,jqfmt=vfsv0,usrjquota=quota.user 0 0
/dev/loop0 /var/suexec/ayannka/chroot/tmp ext4 rw,nosuid,nodev,noexec,noatime,nodiratime,barrier=1,data=ordered 0 0
/dev/sda4 /var/suexec/ayannka/home/ayannka ext4
rw,noatime,nodiratime,barrier=1,data=ordered,jqfmt=vfsv0,usrjquota=quota.user 0 0
If there is a problem, issue '/etc/init.d/suexec-chroot cleanuser ayannka' then check the mounts again
Never mount folder for a user by hand. This is not supported by the suexec-chroot script and if you do not
understand how it works, do not try to mount anything by hand. Let hive_exec or cron do the mounting for you.

2. Check if all base mounts are ok. You should have at least the following 3:
none on /var/suexec/baseos type tmpfs (rw,size=600m,size=600m)
/tmp on /chroot/tmp type none (rw,bind)
/chroot on /var/suexec/baseos/chroot type none (rw,bind)

3. Check if the user limit is not reached. ps axu|grep -c USERNAME and check the result is more then 40. If it is
more then 40 check if the user has specific limit grep USERNAME /usr/local/apache/conf/rlimit-config. Also you can
check for setuid errors in the suexec_log. If the limit is a problem you can either disable it for this user, or
increase it.
Hive limits configuration: http:docs.1h.com/Suexec#Configuration
Apache exclude from limits: http:docs.1h.com/Mod_hive:_options#Disable_user_limitations
Crond exclude from limits: http://docs.1h.com/Crond#Exclude_an_account_from_global_limits