Overcommit warning in the console log, despite vm.overcommit_memory has been se

  • Resolved Carsten Lund

    (@carsten-lund)


    3 years, 2 months ago

    I receive the overcommit warning in the console log, despite vm.overcommit_memory has been set to 1, and rebooted.

    View post on imgur.com

    Any idea why this happens anyway?

    WARNING Memory overcommit must be enabled! Without it, a background save or replication may fail under low memory condition. Being disabled, it can can also cause failures without low memory condition, see https://github.com/jemalloc/jemalloc/issues/1328. To fix this issue add ‘vm.overcommit_memory = 1’ to /etc/sysctl.conf and then reboot or run the command ‘sysctl vm.overcommit_memory=1’ for this to take effect.

Viewing 8 replies - 1 through 8 (of 8 total)
  • Thread Starter Carsten Lund

    (@carsten-lund)

    3 years, 2 months ago

    Hi there, as others have experienced Redis suddenly disconnect to the server. The log reveal problems with missing Handshake with server.

    View post on imgur.com

    Changing the Memory Limit from Unlimited to 2000 mb seems to solve the issues, and the Console Log is loaded is a lot shorter and loads faster.

    The problem is, I really do not now how much memory I should allocate. My server is a VPS S NVMe with 8GB memory.

    Would a quarter of the memory, in this case 2000mb be a good starting point?

    I have been looking for documentation on this topic, but not much guidence.

    Plugin Author Till Krüss

    (@tillkruess)

    3 years, 2 months ago

    I can only help with Redis Object Cache, not with Redis Server maintenance. I’d suggest asking on Stack overflow.

    Thread Starter Carsten Lund

    (@carsten-lund)

    3 years, 2 months ago

    Thanks

    Thread Starter Carsten Lund

    (@carsten-lund)

    3 years, 2 months ago

    Fair enough that you only support Redis Object Cache issues, but as user it can be difficult to distinguish between plugin related errors or server related errors. All you know is that the plugin is disconnected to the server again like in this case.

    1:S 17 Mar 2023 13:47:34.953 * Connecting to MASTER XXXXX
    1:S 17 Mar 2023 13:47:34.954 * MASTER <-> REPLICA sync started
    1:S 17 Mar 2023 13:47:35.033 # Error condition on socket for SYNC: Connection refused

    Plugin Author Till Krüss

    (@tillkruess)

    3 years, 2 months ago

    That error is unrelated to Redis Object Cache. It hints at your Redis replication (more than one server). This is quite advanced. I’d suggest asking your hosting provider or hiring a professional with this.

    Thread Starter Carsten Lund

    (@carsten-lund)

    3 years, 2 months ago

    Thanks, expected it to be advanced to solve, would deleting the instance, and reinstall help?

    Could you help me with a command which delete all Redis content from the server, in case that the instance is installed more than once, as your reply implies, even though it is out of your scope?

    Thanks

    Plugin Author Till Krüss

    (@tillkruess)

    3 years, 2 months ago

    https://redis.io/commands/flushall/

    Thread Starter Carsten Lund

    (@carsten-lund)

    3 years, 2 months ago

    Thanks for you quick reply 😉

    Would it be sync or async?

Viewing 8 replies - 1 through 8 (of 8 total)

The topic ‘Overcommit warning in the console log, despite vm.overcommit_memory has been se’ is closed to new replies.