Memory leak on Panorama from reportd process

• Reportd.log shows an "error handling report request" :
  

  Panorama> grep mp-log reportd.log* pattern "Error handling report request"
  2025-06-26 16:00:15.517 -0400 Error:  reportd_handle_xml_req(req.c:2067): Error handling report request in:reportd
  

• Out-of-memory (oom) error seen in messages log showing "invoked oom-killer":
  

  Panorama> grep mp-log messages pattern "oom"
  Sep 23 10:52:12 panorama kernel: invoked oom-killer: gfp_mask=0x6200ca(GFP_HIGHUSER_MOVABLE), order=0, oom_score_adj=0
  

• Reportd process shows a consistent increase in RES memory when checked over the course of multiple days:
  

  Panorama> show system resources | match "PID\|reportd"
   PID  USER  PR   VIRT    RES    SHR  S %CPU %MEM TIME+    COMMAND
  16539  20    0  2307340 745932 40912 S  0.0  1.1 47:12.00 reportd

  The expected value of reportd RES mem should be under 20GB (20,000,000 in the output above)

• Panorama
• PAN-OS 10.2+
• Strata Logging Service scheduled reports enabled:
  

  Panorama> show system state | match sched-report
  lcass-sched-reports-enabled: True

   

• Inline cloud report activity:
  

  Panorama>  grep mp-log reportd.log* pattern "inline cloud report"
  2025-07-14 00:56:03.086 -0400 Reportjob manager processing inline cloud report results for reportid

The root cause is a memory leak in the reportd process. This issue is identified as PAN-296635.

1. Upgrade Panorama to a fixed version of PAN-OS:
For PAN-OS 10.2, the fix is available in 10.2.17+
For PAN-OS 11.2, the fix will be available in 11.2.11+

2. As a temporary workaround, restart the reportd process with the command 'debug software restart process reportd' to bring memory usage back to a normal level.