erratic accounting updates after 47 days of uptime

PPPoE related questions
hashbang
Posts: 135
Joined: 12 Jul 2015, 10:28

Re: erratic accounting updates after 47 days of uptime

Post by hashbang »

ty,
will do it later on as I'd outage last night and update

thanks
hashbang
Posts: 135
Joined: 12 Jul 2015, 10:28

Re: erratic accounting updates after 47 days of uptime

Post by hashbang »

state: active
fail count: 0
request count: 0
queue length: 0
auth sent: 57142
auth lost(total/5m/1m): 22/0/0
auth avg query time(5m/1m): 411/376 ms
acct sent: 11726
acct lost(total/5m/1m): 2433/0/0
acct avg query time(5m/1m): 35/23 ms
interim sent: 0
interim lost(total/5m/1m): 0/0/0
interim avg query time(5m/1m): 0/0 ms
again happened plz see the interim sent 0. Its been 25 minutes gone and interval is 600
logs at https://pastebin.com/mf0cqMSL
thanks
dimka88
Posts: 866
Joined: 13 Oct 2014, 05:51
Contact:

Re: erratic accounting updates after 47 days of uptime

Post by dimka88 »

Hi, do you have defined acct-interim-interval=600 in [radius] radius section? Or can you confirm that radius sends this attribute?
You can join to our telegram international community group https://t.me/accel_ppp_en for fast help.
hashbang
Posts: 135
Joined: 12 Jul 2015, 10:28

Re: erratic accounting updates after 47 days of uptime

Post by hashbang »

hi,
TY,
Its same in radius and accel 600s. Please see the missing updates in graph between 15:00 to 16:00 and reboot around 13:45 hours
Screenshot from 2020-05-23 18-31-38.png
Screenshot from 2020-05-23 18-31-38.png (179.96 KiB) Viewed 4187 times
thanks
dimka88
Posts: 866
Joined: 13 Oct 2014, 05:51
Contact:

Re: erratic accounting updates after 47 days of uptime

Post by dimka88 »

Can I propose to increase log level and set additional verbose?
Add

Code: Select all

[ppp]
verbose=1

[pppoe]
verbose=1

[radius]
verbose=1
interim-verbose=1

[log]
level=5
log-debug=/var/log/accel-ppp/debug.log
After these changes just execute `accel-cmd reload` to apply the configuration.

UPD:/ Provide please `dmesg -T` and journalctl output from 15 to 16
FYI: `interim sent: 0` this is not normal. Is accel-cmd answered when it happened?
hashbang
Posts: 135
Joined: 12 Jul 2015, 10:28

Re: erratic accounting updates after 47 days of uptime

Post by hashbang »

Hi,
TY,
I'd posted the dmesg output in my other post. Here is the link
https://gofile.io/d/kWlaFI
thanks
hashbang
Posts: 135
Joined: 12 Jul 2015, 10:28

Re: erratic accounting updates after 47 days of uptime

Post by hashbang »

Hi,
Plz find link of dmesg and journal of 23rd may
https://gofile.io/d/Ep8Xzz
thanks
dimka88
Posts: 866
Joined: 13 Oct 2014, 05:51
Contact:

Re: erratic accounting updates after 47 days of uptime

Post by dimka88 »

Community members in our telegram chat recommend disabling SELinux.
hashbang
Posts: 135
Joined: 12 Jul 2015, 10:28

Re: erratic accounting updates after 47 days of uptime

Post by hashbang »

dimka88 wrote: 24 May 2020, 18:50 Community members in our telegram chat recommend disabling SELinux.
ty,
its already disabled
hashbang
Posts: 135
Joined: 12 Jul 2015, 10:28

Re: erratic accounting updates after 47 days of uptime

Post by hashbang »

hi,
Again interim updates stopped. I'd reduced the log levels as no resolution was coming. It was consuming my disk space. Restarting the services (not accel-cmd) could only resolve the problem. Acce-cmd restart only gave me a stuck service with thousand of finishing connections and nothing active
screenshot
Attachments
Screenshot from 2020-10-05 23-10-00.png
Screenshot from 2020-10-05 23-10-00.png (175.78 KiB) Viewed 3720 times
Post Reply