high latency network environment - only the instance works
Hi all,
I have a high latency problem:
- when an instance works with xtrabackup or it is a replica and the apps requires ~ 2k read queries, the ping is high
- when an instance doesn't work, and it is only a replica, the ping is low
I've tried some test in an instance, replica of our master with write requests average of 440:
- when the xtrabackup works (it writes in an external HD mounted as a local directory), the ping is
-- x.x.x.34 ping statistics ---
939 packets transmitted, 939 received, 0% packet loss, time 938430ms
rtt min/avg/max/mdev = 0.178/4.
-- I tested a script with 5k queries and the time average results about 10 execution are 10 seconds
- when a moment when the xtrabackup doesn't work, the ping is
-- x.x.x.34 ping statistics ---
332 packets transmitted, 331 received, 0% packet loss, time 331128ms
rtt min/avg/max/mdev = 0.162/0.
-- I tested a script with 5k queries and the time average results about 10 execution are 2 seconds
I've tried some test in our master:
- when the master serves a write requests average of 440, the ping is
-- x.x.x.33 ping statistics ---
859 packets transmitted, 859 received, 0% packet loss, time 858445ms
rtt min/avg/max/mdev = 0.147/2.
-- I tested a script with 5k queries and the time average results about 10 execution are 7.5 seconds
- when the master is without requests
-- x.x.x.33 ping statistics ---
327 packets transmitted, 327 received, 0% packet loss, time 326123ms
rtt min/avg/max/mdev = 0.158/0.
-- I tested a script with 5k queries and the time average results about 10 execution are 1.8 seconds
Versions used:
- Centos 7.3.1611
- Percona Server 5.6.32-78.1-log, Revision 4c779b7
What can I test to understand if the problem is the Percona Server configuration?
Thank you.
Question information
- Language:
- English Edit question
- Status:
- Expired
- Assignee:
- No assignee Edit question
- Last query:
- Last reply: