LiteSpeed is Running but systemd Status Inactive

Sometimes, you may experience a weird error. LiteSpeed processes are running, but system status shows inactive.

ps aux|grep http
root      8806  0.2  0.1 190080 143460 ?       S    03:58   0:36 litespeed (lshttpd)
root      8809  0.0  0.0  18256   568 ?        S    03:58   0:00 httpd (lscgid)
nobody    8847  0.5  0.1 208280 144056 ?       Sl   03:58   1:09 litespeed (lshttpd)
nobody    8848  5.1  0.3 396704 224876 ?       Sl   03:58  10:50 litespeed (lshttpd)
systemctl status lshttpd.service 
● lshttpd.service - LiteSpeed HTTP Server
 Loaded: loaded (/etc/systemd/system/lshttpd.service; enabled; vendor preset: disabled)
 Active: failed (Result: exit-code) since Thu 2018-02-22 03:55:26 CET; 3h 35min ago
 Main PID: 29752 (code=killed, signal=KILL)
 CGroup: /system.slice/lshttpd.service

Sometimes it may be caused by a shared memory issue.

[/dev/shm/lsws]$ ls -alt ; date
total 3536
-rw------- 1 nobody nobody  147456 Feb 23 04:58 .store.0.shm
-rw-r----- 1 root   nobody 2637824 Feb 23 04:06 .quicshmdata.shm
-rw-r----- 1 root   nobody    8192 Feb 23 04:01 .quicshmmap.lock
-rw-r----- 1 root   nobody    8192 Feb 23 04:01 .quicshmdata.lock
-rw------- 1 nobody nobody  262144 Feb 22 23:02 .store.1.shm
-rw------- 1 nobody nobody  212992 Feb 22 20:32 .store.2.shm
-rw------- 1 nobody nobody   98304 Feb 21 14:49 .store.3.shm
-rw------- 1 nobody nobody    8192 Feb 21 07:40 .cachemanidx.lock
-rw------- 1 nobody nobody  147456 Feb 15 01:59 .cachemanidx.shm
-rw-r----- 1 root   nobody   32768 Feb 12 20:54 .quicshmmap.shm
-rw------- 1 nobody nobody    8192 Feb 10 00:29 .store.2.lock
-rw------- 1 nobody nobody    8192 Feb 10 00:28 .store.1.lock
-rw------- 1 nobody nobody    8192 Feb 10 00:23 .store.3.lock
-rw------- 1 nobody nobody    8192 Feb 10 00:22 .store.0.lock
drwxr-x--- 2 nobody nobody     360 Feb 10 00:22 ./
-rw------- 1 nobody nobody    8192 Feb 10 00:22 adns_cache.lock
-rw------- 1 nobody nobody   16384 Feb 10 00:22 adns_cache.shm
drwxrwxrwt 6 root   root       500 Jan  5 00:20 ../

In this case, it seems that some old files dated “Feb 10” are still there. Here's how to fix it:

Remove all of the files in the /dev/shm/lsws directory and restart the server with the following commands:

rm -rf /dev/shm/lsws
service lsws restart

The issue is resolved and systemd status is correct:

systemctl status lshttpd.service
● lshttpd.service - LiteSpeed HTTP Server
 Loaded: loaded (/etc/systemd/system/lshttpd.service; enabled; vendor preset: disabled)
 Active: active (running) since Fri 2018-02-23 07:35:46 CET; 9s ago
Process: 1620 ExecStart=/usr/local/lsws/bin/lswsctrl start (code=exited, status=0/SUCCESS)
Main PID: 1639 (litespeed)
 CGroup: /system.slice/lshttpd.service
         ‣ 1639 litespeed (lshttpd)
 
litespeed_wiki/troubleshooting/litespeed_is_running_but_systemd_status_inactive.txt · Last modified: 2018/02/28 21:09 by Lisa Clarke