| Solution |
After upgrading to FortiManager 7.2.10, the daemon /bin/ptsessionmgrd CPU usage randomly spikes to 100% and crashes.
To verify the FortiManager is hitting this bug:
exe top -b -n 1
top - 14:14:12 up 1 day, 6:21, 0 users, load average: 1.81, 1.75, 1.55
Tasks: 220 total, 3 running, 215 sleeping, 0 stopped, 2 zombie
%Cpu(s): 13.1 us, 11.1 sy, 0.0 ni, 73.7 id, 0.0 wa, 0.0 hi, 2.0 si, 0.0 st
MiB Mem : 32125.9 total, 12470.8 free, 17024.4 used, 2630.7 buff/cache
MiB Swap: 2048.0 total, 2.8 free, 2045.2 used. 14650.3 avail Mem
PID USER PR NI VIRT RES %CPU %MEM TIME+ S COMMAND
20278 root 20 0 67.3m 11.3m 100.0 0.0 1191:18 R /bin/ptsessionmg+ <-----
1052 root 20 0 227.9m 93.8m 20.0 0.3 81:46.26 R fds_worker
9720 apache 20 0 0.0m 0.0m 13.3 0.0 0:00.04 Z [httpd] <defunct>
1025 root 20 0 533.5m 409.3m 6.7 1.3 26:03.60 S fds_svrd
diag debug crash read
2025-02-20 18:04:24 Firmware FMG-VM64 v7.2.10-build1682 250211 (GA)
2025-02-20 18:04:24 Application ptsessionmgrd
2025-02-20 18:04:24 Signal 6 (Aborted)
2025-02-20 18:04:24 Pid 950 ppid 1 backtrace:
2025-02-20 18:04:24 [0x7f57a5f1731c] libsegfault.so +0x131c print_trace+0x37/0x179
2025-02-20 18:04:24 [0x7f57a5f17506] libsegfault.so +0x1506 catch_segfault+0xa8/0x120
2025-02-20 18:04:24 [0x7f57a343b910] libc.so.6 +0x38910
2025-02-20 18:04:24 [0x7f57a343b891] libc.so.6 +0x38891 gsignal+0x141/0x179
2025-02-20 18:04:24 [0x7f57a342555b] libc.so.6 +0x2255b abort+0x127/0x21c
2025-02-20 18:04:24 [0x7f57a347e0d8] libc.so.6 +0x7b0d8
2025-02-20 18:04:24 [0x7f57a348544a] libc.so.6 +0x8244a
2025-02-20 18:04:24 [0x7f57a348726d] libc.so.6 +0x8426d
2025-02-20 18:04:24 [0x562bfdb779e6] ptsessionmgrd +0x59e6
2025-02-20 18:04:24 [0x562bfdb769ce] ptsessionmgrd +0x49ce main+0x28e/0x2b8
2025-02-20 18:04:24 [0x7f57a3426deb] libc.so.6 +0x23deb __libc_start_main+0xeb/0x1be
2025-02-20 18:04:24 [0x562bfdb76aca] ptsessionmgrd +0x4aca
2025-02-20 18:04:24
This is a well-known issue on FortiManager 7.2.10 fixed on 7.4.7 and 7.6.4.
|
