For more information regarding the security incident at F5, the actions we are taking to address it, and our ongoing efforts to protect our customers, click here.

Forum Discussion

Kai_M__48813's avatar
Jan 14, 2015

ASM out of memory error, event code p120

i get this message on my 4200 bigip 11.4.1 appliance. i got a HA pair in active/standby mode. when searching on devcentral and askf5, i find nothing regarding this error message.

 

I assume that a reboot of the appliances will solve this issue, at least temporary. But we are facing with some instabilities on some services published, so any help/input woul be appriciated.

 

5 Replies

  • if you run out of memory, you have 3 options:

     

    1. disable another module like AVR

     

    2. custom memory settings with cli

     

    3. buy a new system with more memory :-)

     

    At resource provisioning, you can see, how much memory is reserved. ASM should run with Nominal setting. with cli and the command top you can look, who much memory is used by the bd process (ASM). If the system is swapping, you need more memory for ASM.

     

    If you are running a lot of xml traffic, you should think about a higher value of total_xml_value. Depending on your release, you can see messages about that in /var/log/ts/bd.log. But F5 removed the log entries anywhere after release 11.2.0 and I didn't find it at 11.5 again.

     

    At the end, you could ask the support for p120.

     

  • thanks for the update. it also appears that there is a known weakness in 11.4.1 regarding memory leaks, that asm module runs out of memory when under ddos attack. this is what happened here, so our next step would be a software upgrade.

     

  • Hi there,

     

    I'm facing the same situation on 6900 HW running 11.3. I got the P120 message a month ago and now face X111 and X239 error codes.

     

    I'm removing AVR module to same a bit of memory but it does not get a lot of fresh air (from 1456MB to 1738MB). Any other hint that does not require purchase a new expensive HW? :)

     

    NB: of course, TMOS upgrade to 11.5 is already in the pipe.

     

    Thanks a lot!

     

  • Hello again,

     

    While investigating these two error codes with the support I noticed the P120 code on another big-IP instance.

     

    Strangely, this instance (6900 with LTM+APM+ASM) does not report a lot of used memory (45MB only):

     

    [root@my_big-ip:Active:In Sync] config grep "xml memory" /var/log/ts/bd.log | tail -n 1 BD_XML|NOTICE|Apr 24 04:44:05.110|25349|xml_validation.cpp:1446|after create of profile 80. (xml memory 46723639 bytes)

     

    On this instance AVR is not provisioned and LTM/APM/ASM are in Nominal mode. total_xml_memory variable is set to default value (0 => 300MB). It's weird to see this "ASM out of memory" error log, don't you think so?

     

    Any hint/guidance will be appreciated.

     

    Thank you