If this is your first visit, be sure to check out the FAQ by clicking the link above. You may have to REGISTER before you can post. To start viewing messages, select the forum that you want to visit from the selection below.
OK. I'm doing it atm. I found one interesting query on proxy db:
delete from proxy_history where id<1425637909 and (clock<1467809065 or (id<=1425637147 and clock<1467809065))
May I disable historySyncer because I've partitioned proxy_history table and created script for drop partitions older than 14 days and how, do you know? I think that it'll solve my problem...
No. HistorySyncer is core application of Zabbix (Server and Proxy)
It updates Configuration cache and write collected data from History and Trend cache in to the database.
Housekeeper process takes care about DB cleaning and can be disabled then partitioning is implemented.
Br, Alexander
Proxy have History tables only, so can switch Housekeeper process off.
Can be done by setting zabbix_proxy.conf parameter:
HousekeepingFrequency=0
and restart of proxy process.
Br, Alexander
The main problem is bigger proxy of two, definitely because all of graphs and data on zzabbix server and smaller proxy are here. But when I put HousekeepingFrequency=0 zabbix proxy cannot be started, I tried that already..
May I disable housekeeper process on zabbix proxy? If I put HousekeepingFrequency=0 in zabbix_proxy.conf it cannot be started (message in log: wrong value of "HousekeepingFrequency" in config file "/etc/zabbix/zabbix_proxy.conf", line 394
But in config file is written:
With HousekeepingFrequency=0 the housekeeper can be only executed using the runtime control option.
Version is 3.0.3. I cannot understand why it isn't working. On the server housekeeper is successfully disabled. I 've changed database on last Friday (installed new one) and now it seems good. Graphs are here:
Hi!
Good! In next step I just want to recommend to reinstall the Proxy
Looks like Proxy anyway have big Queue. Can you check, what it is in Administration - Queue - Overview by Proxy and Details?
Br, Alexander
Thanks for reply, again. Yes in queue are a lot of items delayed over 10 minutes (969 at the moment). One strange thing, all of them are "Description" and "Alias of interface" on diferent snmp devices...
Text types of Items is taking much more load than numeric.
As well, usually "Description" and "Alias of interface" is not changed very often, so may be just set check period to once a day or a week, if it is not yet done.
Br, Alexander
Comment