Redis 5.0.7 restarting

[abhilash p]

Hi,

Redis server gets restarted automatically .Trying to find reasons for restart

[Itamar Haber]

Hello Abhilash,

Is there anything in the logs?

On Wed, Dec 18, 2019 at 8:36 PM abhilash p <26.ab...@gmail.com> wrote:

Hi,

Redis server gets restarted automatically .Trying to find reasons for restart

--
You received this message because you are subscribed to the Google Groups "Redis DB" group.
To unsubscribe from this group and stop receiving emails from it, send an email to redis-db+u...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/redis-db/d0f9f166-21d8-43c4-93bb-f9386698ec3f%40googlegroups.com.

--

Itamar Haber Technicalist Evangely Phone: +972.54.567.9692

[abhilash p]

Hi,

So we have multiple issues here .We are using redis 5.0.1

Setup:

1 master 4 slaves

Sentinel is also configured

In the first instance 

1)one of the redis slaves was down due to server reboot 

2)we observed that max client which is set to 15000 was reached on the remaining master and slaves 

3)Application got hung due to this and also we were unable to login to redis 

4)However when we got the other slave to run the client connections got reduced to ~1200 on each of the servers.

In the second instance :

1)All redis servers including master got restarted on it's own 

2)In the logs it showed user requested shutdown 

3)However all servers got restarted at the exact same time.Hence it's unlikely that someone shut it down .

Thanks and Regards 

Abhilash Pillai 

To view this discussion on the web visit https://groups.google.com/d/msgid/redis-db/CAFWLiegWQnyiiA7wOPoAAHEPtWwxQwtdG92dxsFoRy-KkcFokw%40mail.gmail.com.

[Greg Andrews]

Look at your Linux OS logs to see if the kernel killed the Redis processes due to excessive memory consumption.  These log entries typically have "OOM" (meaning Out Of Memory) in them.

Systemd is often configured to perform health checks on software daemons that it launches, and if your Redis installation has that kind of Systemd configuration, and the overload in connections caused all your Redis daemons to slow down or stop accepting new connections, they may have been killed and re-started by Systemd on the servers.  Check the Systemd logs to confirm/deny this possibility.

To unsubscribe from this group and stop receiving emails from it, send an email to redi...@googlegroups.com.

To view this discussion on the web visit https://groups.google.com/d/msgid/redis-db/d0f9f166-21d8-43c4-93bb-f9386698ec3f%40googlegroups.com.

--

Itamar Haber Technicalist Evangely Phone: +972.54.567.9692

--
You received this message because you are subscribed to the Google Groups "Redis DB" group.

To unsubscribe from this group and stop receiving emails from it, send an email to redi...@googlegroups.com.

[abhilash p]

Hi,

Please find the logs attached.Did not get any memory error .

To unsubscribe from this group and stop receiving emails from it, send an email to redis-db+u...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/redis-db/2aece410-8d7a-4472-b1b4-1aa771c35d22%40googlegroups.com.