I have this problem where Apache starts before NSS is fully loaded and fails as my document directory is on an NSS drive. When I look at the services file it does look like it should not start before nss, but it does. Any ideas? This os OES 2018 which is SLES 12 based.
argo:~ # systemctl list-units | grep nss
media-nss-VOL1.mount loaded active mounted /media/nss/VOL1
opt-novell-nss-mnt-.pools-VOL1_POOL.mount loaded active mounted /opt/novell/nss/mnt/.pools/VOL1_POOL
ncp2nss.service loaded active running Novell NCP/NSS IPC daemon
novell-nss.service loaded active running Novell Storage Services (NSS) file system.
nss-lookup.target loaded active active Host and Network Name Lookups
nss-user-lookup.target loaded active active User and Group Name Lookups
argo:~ #
Should I have it rely on media-nss-VOL1.mount instead?
argo:~ # systemctl list-units | grep nss
media-nss-VOL1.mount loaded active mounted /media/nss/VOL1
opt-novell-nss-mnt-.pools-VOL1_POOL.mount loaded active mounted /opt/novell/nss/mnt/.pools/VOL1_POOL
ncp2nss.service loaded active running Novell NCP/NSS IPC daemon
novell-nss.service loaded active running Novell Storage Services (NSS) file system.
nss-lookup.target loaded active active Host and Network Name Lookups
nss-user-lookup.target loaded active active User and Group Name Lookups
argo:~ #
Should I have it rely on media-nss-VOL1.mount instead?[/QUOTE]
Hi
Looks like it, so if you copy the /usr/lib/systemd/system/apache2.service over to /etc/systemd/system directory and stop/disable the original one, then modify/start/enable the version in /etc/systemd/system/ you should be good to go…
OK. But would not changing:
After=network.target nss-lookup.target time-sync.target remote-fs.target
to:
After=network.target media-nss-VOL1.mount time-sync.target remote-fs.target
[QUOTE=AndersG;58790]OK. But would not changing:
After=network.target nss-lookup.target time-sync.target remote-fs.target
to:
After=network.target media-nss-VOL1.mount time-sync.target remote-fs.target
Work as well?[/QUOTE]
Hi
Yes it would, but the next update and the service gets overwritten… probably rename the service to say nss_apache2.service and disable the apache2 one…
Just check the remote-fs.target as well, that may be where the problem lies?