Options
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
100% CPU consumed because of cw_acd service
Hi,
Since we upgraded our Fortigate 200B cluster to version 5 patch 4 from version 4 MR3 patch 12, after about a week of uptime the cpu goes to 100%. The diag sys top command shows that the cw_acd process is using all the cpu. Doing a exec wireless-controller restart-acd command has no effect. We have to restart the whole machine.
This has happened twice now with a week in between.
9 REPLIES 9
Options
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
This just happened to us on a FortiGate 100D and made our FortAP' s not work.
Found that the cw_acd process was using 100% and using the execute wireless-controller restart-acd command did nothing.
I used the diagnose sys kill 1 <pid> to kill the process and that fixed it.
Options
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
Same problems here, have made 2 cases already 3 weeks ago, still no solution...
[link]https://www.secure-it.be[/link]
[link]https://www.secure-it.be[/link]
Options
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
The kill command did not work in my case, had to restart the fw.
The problem is not fixed in version 5 patch 5. In fact it is worse. Even with background scanning disabled the problem appears after a few days of uptime.
Options
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
Stil no solution from Fortinet, they just keep asking to provide the diagnose output :-s.
i have now completely disabled the webfiltering feature, not sure if that helps solving the wireless outage..
Unbelievable how this issue is taking so long time to resolve and response..
[link]https://www.secure-it.be[/link]
[link]https://www.secure-it.be[/link]
Options
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
you can try # fnsysctl kill -9 <PID> to kill the process as well. Is the process listed as Dead?
If the process is dead it cannot be restarted via a kill signal
Options
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
The diag sys kill 9 command seems to work now on version 5.0.5. Still the problem is worse since it now happens several times a day and still nowhere near a solution.
I got a reply from Fortinet support that the problem has been escalated, after 1 and a half months. This was a week ago and still no update on the problem.
Options
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
Had this same issue on a new 80C with firmware on 5.0.6; i resolved the issue so far by disabling the WIPS
Options
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
We' ve been on v5.0.5 for a few months now without issue. We were not using WIDS profile on our APs. We decided to try it out and less than 24hrs later we experienced the same issue.
FG 200B, FAP220B
Options
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
Please check the kb : https://community.fortinet.com/t5/FortiAP/Troubleshooting-Tip-cw-acd-process-spikes-to-99-on-FortiGa...
PIYUSH PRINCE