Uploaded image for project: 'Red Hat 3scale API Management'
  1. Red Hat 3scale API Management
  2. THREESCALE-12258

API request returning large response fails when using proxy service policy

XMLWordPrintable

    • Icon: Bug Bug
    • Resolution: Unresolved
    • Icon: Major Major
    • None
    • 2.15.3 GA, 2.15.5 GA, 2.16.1 GA
    • Gateway
    • False
    • Hide

      None

      Show
      None
    • False
    • Not Started
    • Not Started
    • Not Started
    • Not Started
    • Not Started
    • Not Started
    • Workaround Exists
    • Hide

      Increase apicast's memory limit as much as possible. 

      Show
      Increase apicast's memory limit as much as possible. 
    • Important

      When using the Proxy Service policy (http_proxy), API requests returning a large response (e.g., 15MB) fail with a 502 Bad Gateway error due to worker process termination. The following error then appears in the APIcast log:

      2026/01/29 03:02:53 [notice] 14#14: signal 17 (SIGCHLD) received from 24
2026/01/29 03:02:53 [alert] 14#14: worker process 24 exited on signal 9
2026/01/29 03:02:53 [notice] 14#14: start worker process 41
2026/01/29 03:02:53 [notice] 14#14: signal 29 (SIGIO) received

      In the worker node where the apicast pod was running, dmesg -T command show the following message.

      [Thu Jan 29 03:01:41 2026] Memory cgroup out of memory: Killed process 2769202 (openresty) total-vm:295000kB, anon-rss:120944kB, file-rss:128kB, shmem-rss:512kB, UID:1000690000 pgtables:416kB oom_score_adj:-997

      This issue does not happen when proxy service policy is not used.

              rhn-support-atra An Tran
              rhn-support-hmiura Hiroko Miura
              Votes:
              0 Vote for this issue
              Watchers:
              3 Start watching this issue

                Created:
                Updated: