Il 23/06/2015 16:09, Jérôme BERTHIER ha scritto:
Hi,
Hi Jerome,
First of all, I would thank you very much for this great tool. ntopng is more
and more impressive after each update !
Few days ago, I upgraded from ntopng 1.2 to 2.0 using yum on Cent0S 7.
I have noticed few issues :
1) the file ntopng.start was erased during the update process. The service went
down. I took a long time before thinking to create it again :-) .
2) ntopng.service status records a lot of message about rm :
Jun 23 15:07:40 myserver.mydomain.tld ntopng[4629]: /bin/rm: missing operand
Jun 23 15:07:40 myserver.mydomain.tld ntopng[4629]: Try '/bin/rm --help' for
more information.
=> I noticed that the exec script searches only for "-G" option to find the pid
file but I set the "--pid" option. So, I modified the script this way :
#PID_FILE=$(cat /etc/ntopng/ntopng.conf | grep '\-G='|cut -d '=' -f 2)
PID_FILE=$(cat /etc/ntopng/ntopng.conf | grep -E "\-G=|--pid"|cut -d '=' -f 2)
3) last issue, not resolved for now, if I reboot the server, ntopng starts with
errors. The web UI is up but no data is available. If I restart the ntop
service, everything goes OK. :
Jun 23 15:00:23 myserver.mydomain.tld ntopng[3154]: [src/Redis.cpp:108] ERROR:
LOADING Redis is loading the dataset in memory
Jun 23 15:00:23 myserver.mydomain.tld ntopng[3154]: [src/Redis.cpp:1297] ERROR:
LOADING Redis is loading the dataset in memory
Jun 23 15:00:23 myserver.mydomain.tld ntopng[3154]: [src/Redis.cpp:125] ERROR:
LOADING Redis is loading the dataset in memory
Jun 23 15:00:23 myserver.mydomain.tld ntopng[3154]: [src/Redis.cpp:108] ERROR:
LOADING Redis is loading the dataset in memory
Have you tried to reset Redis with redis-cli flushall?
=> so I tried to modify the systemd unit ntopng.service to set it to start after
redis.service.
In file /etc/systemd/system/ntopng.service, add "redis.service" to the "After"
line :
That is a great idea, thanks for suggesting it.
Arianna
After=network.target syslog.target redis.service
At next reboot, ntopng correcty start after redis but the error is still there :
[root@myserver ~]# systemd-analyze critical-chain
The time after the unit is active or started is printed after the "@" character.
The time the unit takes to start is printed after the "+" character.
multi-user.target @19.654s
└─ntopng.service @15.550s +4.103s
└─redis.service @15.550s
└─network.target @15.534s
Any ideas ?
Regards,
_______________________________________________
Ntop mailing list
[email protected]
http://listgateway.unipi.it/mailman/listinfo/ntop
--
/*
* Arianna Avanzini
* [email protected]
* http://ava.webhop.me
*/
_______________________________________________
Ntop mailing list
[email protected]
http://listgateway.unipi.it/mailman/listinfo/ntop
