mysql connection,network usage hourly ,select type can’t show in pmm version 1.1.5,others is normal.
can you help me to check ?
kaky2000
my apologies for the delay on this topic.
how many pmm clients do you have?
can you check if data for “MySQL Network Usage Hourly” graph exists in prometheus directly?
it is needed to open this link [url]http://PMM-SERVER-IP/prometheus/graph?g0.range_input=1h&g0.expr=mysql_global_status_bytes_received&g0.tab=0[/url]
can you share a screenshot?
TOTOL 6 clients,3 for mysql,3 for mongodb.
1、can you check if data for “MySQL Network Usage Hourly” graph exists in prometheus directly? could you give me a guideline ? 2、it is needed to open this link [url]http://PMM-SERVER-IP/prometheus/grap...eived&g0.tab=0[/url]
it is possible to do it manually:
0. add “prometheus/” to the end of the PMM Server URL (like [url]http://PMM-SERVER-IP/prometheus/[/url])
- open “Graph” page
- enter “mysql_global_status_bytes_received” metric name to “Expression” field
- click “Execute” button
- choose “Graph” tab on “Graph” page
- increase the period to 12h until now.
see find the screenshot below.
can you check the following expressions?
max(max_over_time(mysql_global_status_threads_connected[24h]))
and
increase(mysql_global_status_bytes_received[1h])
I need load time details for both expressions.
could you help me check?
kaky2000, my apologies for the delay on this topic.
how many RAM do you have?
can you increase prometheus memory up to 8Gb ? METRICS_MEMORY=5991800
see [URL=“Percona Monitoring and Management”]https://www.percona.com/doc/percona-...mption-for-pmm[/URL]
256GB
i create and run the command:(set 50GB)
docker run -d -p 80:80 --volumes-from pmm-data -e METRICS_MEMORY=50194304 -e METRICS_RESOLUTION=5s --name pmm-server --restart always docker.gf.com.cn/percona/pmm-server:1.1.5
But it still slow,check the picture.
Hi kaky2000 ,
it is expectable, prometheus use only needed the amount of RAM
How many pmm-clients do you have?
can you share ?
can you restart prometheus?
docker exec -it pmm-server supervisorctl restart prometheus
is it acceptable to lose all historical data for all graphs?
I try to stop docker,and recreate the docker (readd the client monitor）,it can work normally now，and now it can use more memory .