I have begin to see the dashboard is not updating ina timely manner after a 3.2.11 to 3.4.9.
Some status:
I have a CentOS 7.4 for the Zabbix server and Web GUI and a CentOS 7.4 running a MySQL 5.7 database. Both servers are VM's with 4 vcpu and 16GB ram
There are 490 host to monitor with 41880 items, requiring 361.16 value per sec.
The host status is showing 13 host are unavaliable, when I check these i find only 2 are really unavialable.
When i look at the server performance graphs, all looks good to me, attached is a doc with screen shots. The gap in the graphs are when I did the upgrade.
Graphs.docx
I have use the mysqltune.pl script to adjust the database performance.
I have opened up the zabbix server cache and processes. and it was working with the 3.2.11 correctly.
I have noticed the disk IO on tehdatabase server is a little high, but seems to recover after 2 to 5 mins, then repeats again when the housekeeping kicks off again.
Some zabbix_server.conf items
StartPollers=154
StartPollersUnreachable=48
StartTrappers=56
StartTimers=8
StartEscalators=10
MaxHousekeeperDelete=100000
CacheSize=64M
StartDBSyncers=24
HistoryCacheSize=768M
HistoryIndexCacheSize=96M
TrendCacheSize=128M
ValueCacheSize=128M
Timeout=30
It seems to me that the GUI can not keep up with what is being updated to the database.
Some status:
I have a CentOS 7.4 for the Zabbix server and Web GUI and a CentOS 7.4 running a MySQL 5.7 database. Both servers are VM's with 4 vcpu and 16GB ram
There are 490 host to monitor with 41880 items, requiring 361.16 value per sec.
The host status is showing 13 host are unavaliable, when I check these i find only 2 are really unavialable.
When i look at the server performance graphs, all looks good to me, attached is a doc with screen shots. The gap in the graphs are when I did the upgrade.
Graphs.docx
I have use the mysqltune.pl script to adjust the database performance.
I have opened up the zabbix server cache and processes. and it was working with the 3.2.11 correctly.
I have noticed the disk IO on tehdatabase server is a little high, but seems to recover after 2 to 5 mins, then repeats again when the housekeeping kicks off again.
Some zabbix_server.conf items
StartPollers=154
StartPollersUnreachable=48
StartTrappers=56
StartTimers=8
StartEscalators=10
MaxHousekeeperDelete=100000
CacheSize=64M
StartDBSyncers=24
HistoryCacheSize=768M
HistoryIndexCacheSize=96M
TrendCacheSize=128M
ValueCacheSize=128M
Timeout=30
It seems to me that the GUI can not keep up with what is being updated to the database.