[
https://issues.apache.org/jira/browse/TS-3164?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14356900#comment-14356900
]
Zhaonanli commented on TS-3164:
-------------------------------
I also found that, and at the same time, ET_AIO is cpu 100%, not all. I think
this is triggered by ET_AIO; my perf top is:
28.30% libtsutil.so.4.2.2 [.] ink_atomiclist_pop
13.47% libtsutil.so.4.2.2 [.] bool
ink_atomic_cas<__int128>(__int128 volatile*, __int128, __int128)
9.94% libtsutil.so.4.2.2 [.] unsigned int
ink_atomic_increment<unsigned int, int>(unsigned int volatile
6.32% libtsutil.so.4.2.2 [.] int
ink_atomic_increment<int, int>(int volatile*, int)
4.56% [kernel] [k] _spin_lock
4.34% libtsutil.so.4.2.2 [.] ink_atomiclist_push
3.41% [kernel] [k] rmap_walk
3.19% traffic_server [.] Mutex_trylock(ProxyMutex*,
EThread*)
2.88% libtsutil.so.4.2.2 [.]
get_chunk_info_addr(_InkFreeList*, void*)
2.87% [kernel] [k] try_to_unmap_anon
2.79% libtsutil.so.4.2.2 [.] reclaimable_freelist_new
1.81% libpthread-2.12.so [.] pthread_mutex_trylock
0.91% traffic_server [.] int
ink_atomic_increment<int, int>(int volatile*, int)
0.78% traffic_server [.] EThread::execute()
0.72% traffic_server [.]
Continuation::handleEvent(int, void*)
0.71% traffic_server [.] DLL<Event,
Event::Link_link>::insert(Event*, Event*)
0.68% traffic_server [.] CacheVC::cancel_trigger()
0.67% traffic_server [.] RefCountObj::refcount_inc()
0.62% traffic_server [.]
CacheVC::openReadStartHead(int, Event*)
0.58% traffic_server [.] DLL<Event,
Event::Link_link>::pop()
0.58% libtsutil.so.4.2.2 [.] bool ink_atomic_cas<int>(int
volatile*, int, int)
0.50% traffic_server [.]
Ptr<ProxyMutex>::operator=(ProxyMutex*)
> why the load of trafficserver occurrs a abrupt rise on a occasion ?
> -------------------------------------------------------------------
>
> Key: TS-3164
> URL: https://issues.apache.org/jira/browse/TS-3164
> Project: Traffic Server
> Issue Type: Bug
> Components: Core
> Environment: CentOS 6.3 64bit, 8 cores, 128G mem
> Reporter: taoyunxing
> Fix For: sometime
>
>
> I use Tsar to monitor the traffic status of the ATS 4.2.0, and come across
> the following problem:
> {code}
> Time ---cpu-- ---mem-- ---tcp-- -----traffic---- --sda--- --sdb---
> --sdc--- ---load-
> Time util util retran bytin bytout util util
> util load1
> 03/11/14-18:20 40.67 87.19 3.36 24.5M 43.9M 13.02 94.68
> 0.00 5.34
> 03/11/14-18:25 40.30 87.20 3.27 22.5M 42.6M 12.38 94.87
> 0.00 5.79
> 03/11/14-18:30 40.84 84.67 3.44 21.4M 42.0M 13.29 95.37
> 0.00 6.28
> 03/11/14-18:35 43.63 87.36 3.21 23.8M 45.0M 13.23 93.99
> 0.00 7.37
> 03/11/14-18:40 42.25 87.37 3.09 24.2M 44.8M 12.84 95.77
> 0.00 7.25
> 03/11/14-18:45 42.96 87.44 3.46 23.3M 46.0M 12.96 95.84
> 0.00 7.10
> 03/11/14-18:50 44.00 87.42 3.49 22.3M 43.0M 14.17 94.99
> 0.00 6.57
> 03/11/14-18:55 42.20 87.44 3.46 22.3M 43.6M 13.19 96.05
> 0.00 6.09
> 03/11/14-19:00 44.90 87.53 3.60 23.6M 46.5M 13.61 96.67
> 0.00 8.06
> 03/11/14-19:05 46.26 87.73 3.24 25.8M 49.1M 15.39 94.05
> 0.00 9.98
> 03/11/14-19:10 43.85 87.69 3.19 25.4M 50.9M 12.88 97.80
> 0.00 7.99
> 03/11/14-19:15 45.28 87.69 3.36 25.6M 49.6M 13.10 96.86
> 0.00 7.47
> 03/11/14-19:20 44.11 85.20 3.29 24.1M 47.8M 14.24 96.75
> 0.00 5.82
> 03/11/14-19:25 45.26 87.78 3.52 24.4M 47.7M 13.21 95.44
> 0.00 7.61
> 03/11/14-19:30 44.83 87.80 3.64 25.7M 50.8M 13.27 98.02
> 0.00 6.85
> 03/11/14-19:35 44.89 87.78 3.61 23.9M 49.0M 13.34 97.42
> 0.00 7.04
> 03/11/14-19:40 69.21 88.88 0.55 18.3M 33.7M 11.39 71.23
> 0.00 65.80
> 03/11/14-19:45 72.47 88.66 0.27 15.4M 31.6M 11.51 72.31
> 0.00 11.56
> 03/11/14-19:50 44.87 88.72 4.11 22.7M 46.3M 12.99 97.33
> 0.00 8.29
> {code}
>
> in addition, top command show
> {code}
> hi:0
> ni:0
> si:45.56
> st:0
> sy:13.92
> us:12.58
> wa:14.3
> id:15.96
> {code}
> who help me ? thanks in advance.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
