Diversion CPU dump with diversion standard on RT-AX58U

[jata]

Hi all,

After installing diversion standard on my RT-AX3000 (same as RT-AX58U), I see the following logged every 20mins.

Diversion seems to be working fine.

I don't get this issue if i install 'diversion lite' so i'm guessing it might be related to pixelserv?

Can anyone help with this?

thanks in advance.

Nov 18 12:49:23 kernel: CPU: 1 PID: 21730 Comm: dcd Tainted: P O 4.1.52 #1
Nov 18 12:49:23 kernel: Hardware name: Generic DT based system
Nov 18 12:49:23 kernel: task: d2aa0400 ti: cbe1a000 task.ti: cbe1a000
Nov 18 12:49:23 kernel: PC is at 0xb6c1e39c
Nov 18 12:49:23 kernel: LR is at 0x1dd14
Nov 18 12:49:23 kernel: pc : [<b6c1e39c>] lr : [<0001dd14>] psr: 600f0010
Nov 18 12:49:23 kernel: sp : bed91908 ip : 000a2050 fp : b5eff024
Nov 18 12:49:23 kernel: r10: 000a23c4 r9 : b5eff95c r8 : 000a287c
Nov 18 12:49:23 kernel: r7 : b5eff994 r6 : 000a2876 r5 : 00000000 r4 : b5eff940
Nov 18 12:49:23 kernel: r3 : 00000000 r2 : bed918e4 r1 : 0007d612 r0 : 00000000
Nov 18 12:49:23 kernel: Flags: nZCv IRQs on FIQs on Mode USER_32 ISA ARM Segment user
Nov 18 12:49:23 kernel: Control: 10c5387d Table: 0f49804a DAC: 00000015
Nov 18 12:49:23 kernel: CPU: 1 PID: 21730 Comm: dcd Tainted: P O 4.1.52 #1
Nov 18 12:49:23 kernel: Hardware name: Generic DT based system
Nov 18 12:49:23 kernel: [<c0026e60>] (unwind_backtrace) from [<c0022c38>] (show_stack+0x10/0x14)
Nov 18 12:49:23 kernel: [<c0022c38>] (show_stack) from [<c045237c>] (dump_stack+0x8c/0xa0)
Nov 18 12:49:23 kernel: [<c045237c>] (dump_stack) from [<c003aad4>] (get_signal+0x490/0x558)
Nov 18 12:49:23 kernel: [<c003aad4>] (get_signal) from [<c00221d0>] (do_signal+0xc8/0x3ac)
Nov 18 12:49:23 kernel: [<c00221d0>] (do_signal) from [<c0022658>] (do_work_pending+0x94/0xa4)
Nov 18 12:49:23 kernel: [<c0022658>] (do_work_pending) from [<c001f4cc>] (work_pending+0xc/0x20

 

[jata]

Just adding to the original post as I continue to investigate...

1. the CPU dump / crashed process happens every 25mins
2. memory use on the router is creeping up (approx. 50mb / day) possible memory leak?
3. the PID for the crashed process changes each time

Anyone have any suggestions or advice to help me investigate further?

Below are the log entries for the last 2 crashes...

Nov 19 09:37:52 kernel: CPU: 1 PID: 18773 Comm: dcd Tainted: P O 4.1.52 #1
Nov 19 09:37:52 kernel: Hardware name: Generic DT based system
Nov 19 09:37:52 kernel: task: cde7c000 ti: d0b44000 task.ti: d0b44000
Nov 19 09:37:52 kernel: PC is at 0xb6bd739c
Nov 19 09:37:52 kernel: LR is at 0x1dd14
Nov 19 09:37:52 kernel: pc : [<b6bd739c>] lr : [<0001dd14>] psr: 600f0010
Nov 19 09:37:52 kernel: sp : beaba908 ip : 000a2050 fp : b5eff024
Nov 19 09:37:52 kernel: r10: 000a23c4 r9 : b5eff95c r8 : 000a287c
Nov 19 09:37:52 kernel: r7 : b5eff994 r6 : 000a2876 r5 : 00000000 r4 : b5eff940
Nov 19 09:37:52 kernel: r3 : 00000000 r2 : beaba8e4 r1 : 0007d612 r0 : 00000000
Nov 19 09:37:52 kernel: Flags: nZCv IRQs on FIQs on Mode USER_32 ISA ARM Segment user
Nov 19 09:37:52 kernel: Control: 10c5387d Table: 120f404a DAC: 00000015
Nov 19 09:37:52 kernel: CPU: 1 PID: 18773 Comm: dcd Tainted: P O 4.1.52 #1
Nov 19 09:37:52 kernel: Hardware name: Generic DT based system
Nov 19 09:37:52 kernel: [<c0026e60>] (unwind_backtrace) from [<c0022c38>] (show_stack+0x10/0x14)
Nov 19 09:37:52 kernel: [<c0022c38>] (show_stack) from [<c045237c>] (dump_stack+0x8c/0xa0)
Nov 19 09:37:52 kernel: [<c045237c>] (dump_stack) from [<c003aad4>] (get_signal+0x490/0x558)
Nov 19 09:37:52 kernel: [<c003aad4>] (get_signal) from [<c00221d0>] (do_signal+0xc8/0x3ac)
Nov 19 09:37:52 kernel: [<c00221d0>] (do_signal) from [<c0022658>] (do_work_pending+0x94/0xa4)
Nov 19 09:37:52 kernel: [<c0022658>] (do_work_pending) from [<c001f4cc>] (work_pending+0xc/0x20)


Nov 19 10:04:18 kernel: CPU: 1 PID: 21652 Comm: dcd Tainted: P O 4.1.52 #1
Nov 19 10:04:18 kernel: Hardware name: Generic DT based system
Nov 19 10:04:18 kernel: task: d28a7c00 ti: cf480000 task.ti: cf480000
Nov 19 10:04:18 kernel: PC is at 0xb6cc139c
Nov 19 10:04:18 kernel: LR is at 0x1dd14
Nov 19 10:04:18 kernel: pc : [<b6cc139c>] lr : [<0001dd14>] psr: 600f0010
Nov 19 10:04:18 kernel: sp : bea10908 ip : 000a2050 fp : b5fff024
Nov 19 10:04:18 kernel: r10: 000a23c4 r9 : b5fff95c r8 : 000a287c
Nov 19 10:04:18 kernel: r7 : b5fff994 r6 : 000a2876 r5 : 00000000 r4 : b5fff940
Nov 19 10:04:18 kernel: r3 : 00000000 r2 : bea108e4 r1 : 0007d612 r0 : 00000000
Nov 19 10:04:18 kernel: Flags: nZCv IRQs on FIQs on Mode USER_32 ISA ARM Segment user
Nov 19 10:04:18 kernel: Control: 10c5387d Table: 0d68804a DAC: 00000015
Nov 19 10:04:18 kernel: CPU: 1 PID: 21652 Comm: dcd Tainted: P O 4.1.52 #1
Nov 19 10:04:18 kernel: Hardware name: Generic DT based system
Nov 19 10:04:18 kernel: [<c0026e60>] (unwind_backtrace) from [<c0022c38>] (show_stack+0x10/0x14)
Nov 19 10:04:18 kernel: [<c0022c38>] (show_stack) from [<c045237c>] (dump_stack+0x8c/0xa0)
Nov 19 10:04:18 kernel: [<c045237c>] (dump_stack) from [<c003aad4>] (get_signal+0x490/0x558)
Nov 19 10:04:18 kernel: [<c003aad4>] (get_signal) from [<c00221d0>] (do_signal+0xc8/0x3ac)
Nov 19 10:04:18 kernel: [<c00221d0>] (do_signal) from [<c0022658>] (do_work_pending+0x94/0xa4)
Nov 19 10:04:18 kernel: [<c0022658>] (do_work_pending) from [<c001f4cc>] (work_pending+0xc/0x20)

 

[dave14305]

As a curiosity, you can run this command to mimic dcd with extra debug output to syslog to see why it crashes.

dcd -i 3600 -p 43200 -b -l 8 -d /tmp/bwdpi/

Borrowed from page 2 of this thread:

[Solved] AC86U - possible memory leaks. AiProtection(dcd) crashing is guilty

Hi. I have an issue on my AC86U - after some period of time the cpu consumption by [usb-storage] process start growing After that the dhcp (i.e. dnsmasq) is going to be unresponcible, due to I use diversion, which was dependent on usb performance, where blacklists are located. The simple...

www.snbforums.com

 

[jata]

thanks for the help Dave.

admin@RT-AX3000:/tmp/home/root# dcd -i 3600 -p 43200 -b -l 8 -d /tmp/bwdpi/
dcd[31217]: [init_syslog(41)] syslog is opened for dcd w/ opts=33
dcd[31217]: [parse_cmd(405)] in_background=1
dcd[31217]: [parse_cmd(411)] coll_intl=1800 conf_poll_intr=43200 verbose=1 one-shot=0
dcd[31217]: [parse_cmd(415)] wdir=/tmp/bwdpi/ fbe_wdir=/tmp/bwdpi//tmfbe_workdir fbe_cert_path=/tmp/bwdpi//ntdasus2014.cert
dcd[31217]: [parse_cmd(416)] syslog_pri=6
dcd[31217]: [parse_cmd(420)] ri-buf-size=1 rs-buf-size=2 buf-size=2
non-opt ARGV-elements: 8
dcd[31217]: [main(1017)] pid=31217, pgid=31196, sid=31196
admin@RT-AX3000:/tmp/home/root# dcd[31219]: [main(1027)] another instance is running; exit ...

 

[dave14305]

killall dcd and then immediately run the original command.

 

[jata]

admin@RT-AX3000:/tmp/home/root# killall dcd
dcd[4742]: [handle_sig(540)] catch sig_no=15

admin@RT-AX3000:/tmp/home/root# dcd -i 3600 -p 43200 -b -l 8 -d /tmp/bwdpi/
dcd[4908]: [init_syslog(41)] syslog is opened for dcd w/ opts=33
dcd[4908]: [parse_cmd(405)] in_background=1
dcd[4908]: [parse_cmd(411)] coll_intl=1800 conf_poll_intr=43200 verbose=1 one-shot=0
dcd[4908]: [parse_cmd(415)] wdir=/tmp/bwdpi/ fbe_wdir=/tmp/bwdpi//tmfbe_workdir fbe_cert_path=/tmp/bwdpi//ntdasus2014.cert
dcd[4908]: [parse_cmd(416)] syslog_pri=6
dcd[4908]: [parse_cmd(420)] ri-buf-size=1 rs-buf-size=2 buf-size=2
non-opt ARGV-elements: 8
dcd[4908]: [main(1017)] pid=4908, pgid=4675, sid=4675
admin@RT-AX3000:/tmp/home/root# dcd[4910]: [load_offline_coll_conf(477)] set eula agreed
dcd[4910]: [reg_sig(641)] before pthread_create()
dcd[4910]: [reg_sig(648)] sig thread is created
dcd[4910]: [prep_conf_serv(131)] local socket created w/ sd=4, path=/var/conf_serv_sock
dcd[4910]: [prep_conf_serv(133)] sizeof(conf_store)=3264
dcd[4910]: [start_serv(814)] service thread is created
dcd[4910]: [start_conf_poller(881)] conf poller thread is started
dcd[4910]: [start_pctrl_thread(940)] pctrl thread is started
dcd[4910]: [tmfbe_start_service(217)] TM_FB_initLog() --> [OK]
dcd[4910]: [tmfbe_start_service(222)] TM_FB_initEng() --> [OK]
dcd[4910]: [tmfbe_start_service(229)] TM_FB_allocCtx() --> [OK]
dcd[4910]: [tmfbe_start_service(247)] Feedback server = ntd-asus-2014b-en.fbs20.trendmicro.com:443
dcd[4910]: [tmfbe_start_service(257)] Cert file size = 2524
dcd[4910]: [conf_poll_func(851)] to delay 9016 sec ...
dcd[4910]: [tmfbe_start_service(294)] Generated guid = 2d1dee7f-5369-2052-2d29-9d69e5cdf80d
dcd[4910]: [tmfbe_start_service(301)] write guid to file ...
dcd[4910]: [tmfbe_start_service(373)] TM_FB_setCtxOpt(TM_FB_CTX_OPT_FBS_INFO) --> [OK]
dcd[4910]: [tmfbe_start_service(373)] TM_FB_setCtxOpt(TM_FB_CTX_OPT_CLIENT_CERT) --> [OK]
dcd[4910]: [tmfbe_start_service(373)] TM_FB_setCtxOpt(TM_FB_CTX_OPT_GUID) --> [OK]
dcd[4910]: [tmfbe_start_service(373)] TM_FB_setCtxOpt(TM_FB_CTX_OPT_DNS_TIMEOUT) --> [OK]
dcd[4910]: [tmfbe_start_service(373)] TM_FB_setCtxOpt(TM_FB_CTX_OPT_FBS_CONN_TIMEOUT) --> [OK]
dcd[4910]: [tmfbe_start_service(373)] TM_FB_setCtxOpt(TM_FB_CTX_OPT_FBS_TRANS_TIMEOUT) --> [OK]
dcd[4910]: [tmfbe_start_service(373)] TM_FB_setCtxOpt(TM_FB_CTX_OPT_BATCH_TIME_INTERVAL) --> [OK]
dcd[4910]: [tmfbe_start_service(373)] TM_FB_setCtxOpt(TM_FB_CTX_OPT_BATCH_ENTRY_NUM) --> [OK]
dcd[4910]: [tmfbe_start_service(373)] TM_FB_setCtxOpt(TM_FB_CTX_OPT_PRE_FEEDBACK_FUNC) --> [OK]
dcd[4910]: [tmfbe_start_service(373)] TM_FB_setCtxOpt(TM_FB_CTX_OPT_POST_FEEDBACK_FUNC1) --> [OK]
dcd[4910]: [tmfbe_start_service(379)] TM_FB_startService() --> [OK]
dcd[4910]: [main(1082)] write stat to /tmp/bwdpi//dcd.stat ...
dcd[4910]: [update_tmufe_counters(1001)] remote query count=1434
dcd[4910]: [update_tmufe_counters(1009)] local query count=18226
dcd[4910]: [main(1093)] to delay 1488 sec ...
dcd[4910]: [accept_msg(631)] read 5 bytes

 

[dave14305]

Let it run until it crashes.

 

[jata]

It's been running for 1.5hrs now. Have seen the crash in my syslog but nothing in the shell that is running the dcd command.

What am I looking for?

edit - just noticed that the shell has crashed - happens when the CPU issue occurs so nothing in the output

 

[dave14305]

Like in the other thread there was an enumeration of router interfaces in the log before it segfaulted. I was curious if it would be a similar situation here.

 

[jata]

I see. Thanks for the explanation. There was nothing indicating enumeration in the shell/output before it crashed

 

[jata]

I have removed Diversion and everything back to normal. Maybe I will try a different adblocker solution in the future...

Thanks for trying to help Dave and also massive thanks for the work on FlexQOS. Currently the only addon I am using.