Login | Register For Free | Help
Search for: (Advanced)

Mailing List Archive: Varnish: Bugs

#1051: child process died

 

 

Varnish bugs RSS feed   Index | Next | Previous | View Threaded


varnish-bugs at varnish-cache

Nov 5, 2011, 6:28 AM

Post #1 of 8 (149 views)
Permalink
#1051: child process died

#1051: child process died
---------------------+------------------------------------------------------
Reporter: sreniaw | Type: defect
Status: new | Priority: normal
Milestone: | Component: varnishd
Version: 3.0.2 | Severity: normal
Keywords: |
---------------------+------------------------------------------------------
I am having problem with varnish 64bit on centos 5.5.
Child process died. New process is not respawning.

Can you help ?

{{{
Nov 5 13:37:41 blade414 Condition((const void*)(o) >= (const
void*)((sg->sc)->base) && (const void*)(o) < (const void *)((sg->sc)->base
+ (sg->sc)->mediasize)) not true.
Nov 5 13:37:41 blade414 thread = (cache-worker)
Nov 5 13:37:41 blade414 ident =
Linux,2.6.18-238.el5,x86_64,-smalloc,-spersistent,-smalloc,-hclassic,epoll
Nov 5 13:37:41 blade414 Backtrace:
Nov 5 13:37:41 blade414 0x42c7a6: /usr/sbin/varnishd [0x42c7a6]
Nov 5 13:37:41 blade414 0x448edc: /usr/sbin/varnishd [0x448edc]
Nov 5 13:37:41 blade414 0x425873: /usr/sbin/varnishd(HSH_Lookup+0x3a3)
[0x425873]
Nov 5 13:37:41 blade414 0x41423e: /usr/sbin/varnishd [0x41423e]
Nov 5 13:37:41 blade414 0x417a92: /usr/sbin/varnishd(CNT_Session+0x9d2)
[0x417a92]
Nov 5 13:37:41 blade414 0x42efb8: /usr/sbin/varnishd [0x42efb8]
Nov 5 13:37:41 blade414 0x42e19b: /usr/sbin/varnishd [0x42e19b]
Nov 5 13:37:41 blade414 0x38be20673d: /lib64/libpthread.so.0
[0x38be20673d]
Nov 5 13:37:41 blade414 0x38bdad3f6d: /lib64/libc.so.6(clone+0x6d)
[0x38bdad3f6d]
Nov 5 13:37:41 blade414 sp = 0x2aaf85cbd008 {
Nov 5 13:37:41 blade414 fd = 1692, id = 1692, xid = 518917304,
Nov 5 13:37:41 blade414 client = 1.1.1.1 32755,
Nov 5 13:37:41 blade414 step = STP_LOOKUP,
Nov 5 13:37:41 blade414 handling = hash,
Nov 5 13:37:41 blade414 restarts = 0, esi_level = 0
Nov 5 13:37:41 blade414 flags =
Nov 5 13:37:41 blade414 bodystatus = 4
Nov 5 13:37:41 blade414 ws = 0x2aaf85cbd080 {
Nov 5 13:37:41 blade414 id = "sess",
Nov 5 13:37:41 blade414 {s,f,r,e} = {0x2aaf85cbdc90,+504,+65536,+65536},
Nov 5 13:37:41 blade414 },
Nov 5 13:37:41 blade414 http[req] = {
Nov 5 13:37:41 blade414 ws = 0x2aaf85cbd080[sess]
Nov 5 13:37:41 blade414 "GET",
Nov 5 13:37:41 blade414 "/ph/az/22/22/az222289.jpg",
Nov 5 13:37:41 blade414 "HTTP/1.1",
Nov 5 13:37:41 blade414 "User-Agent: Mozilla/5.0 (Windows NT 5.1;
rv:7.0.1) Gecko/20100101 Firefox/7.0.1",
Nov 5 13:37:41 blade414 "Accept: image/png,image/*;q=0.8,*/*;q=0.5",
Nov 5 13:37:41 blade414 "Accept-Language: pl,en-us;q=0.7,en;q=0.3",
Nov 5 13:37:41 blade414 "Accept-Charset: ISO-8859-2,utf-8;q=0.7,*;q=0.7",
Nov 5 13:37:41 blade414 "Connection: keep-alive",
Nov 5 13:37:41 blade414 "host: static",
Nov 5 13:37:41 blade414 },
Nov 5 13:37:41 blade414 worker = 0x2aaf4ce0ccf0 {
Nov 5 13:37:41 blade414 ws = 0x2aaf4ce0cf30 {
Nov 5 13:37:41 blade414 id = "wrk",
Nov 5 13:37:41 blade414 {s,f,r,e} =
{0x2aaf4cdfaca0,0x2aaf4cdfaca0,(nil),+65536},
Nov 5 13:37:41 blade414 },
Nov 5 13:37:41 blade414 },
Nov 5 13:37:41 blade414 vcl = {
Nov 5 13:37:41 blade414 srcname = {
Nov 5 13:37:41 blade414 "input",
Nov 5 13:37:41 blade414 "Default",
Nov 5 13:37:41 blade414 "/etc/varnish/url.vcl",
Nov 5 13:37:41 blade414 },
Nov 5 13:37:41 blade414 },
Nov 5 13:37:41 blade414 },
Nov 5 13:37:41 blade414 CLI communication error (hdr)
Nov 5 13:37:41 blade414 Condition(sg1->p.offset != sg->p.offset) not
true.
Nov 5 13:37:41 blade414 thread = (cache-main)
Nov 5 13:37:41 blade414 ident =
Linux,2.6.18-238.el5,x86_64,-smalloc,-spersistent,-smalloc,-hclassic,no_waiter
Nov 5 13:37:41 blade414 Backtrace:
Nov 5 13:37:41 blade414 0x42c7a6: /usr/sbin/varnishd [0x42c7a6]
Nov 5 13:37:41 blade414 0x4472f5: /usr/sbin/varnishd [0x4472f5]
Nov 5 13:37:41 blade414 0x4474bb: /usr/sbin/varnishd [0x4474bb]
Nov 5 13:37:41 blade414 0x444d57: /usr/sbin/varnishd(STV_open+0x27)
[0x444d57]
Nov 5 13:37:41 blade414 0x42b525: /usr/sbin/varnishd(child_main+0xc5)
[0x42b525]
Nov 5 13:37:41 blade414 0x43d5ec: /usr/sbin/varnishd [0x43d5ec]
Nov 5 13:37:41 blade414 0x43de7c: /usr/sbin/varnishd [0x43de7c]
Nov 5 13:37:41 blade414 0x3f88c094c7: /usr/lib64/varnish/libvarnish.so
[0x3f88c094c7]
Nov 5 13:37:41 blade414 0x3f88c09b58:
/usr/lib64/varnish/libvarnish.so(vev_schedule+0x88) [0x3f88c09b58]
Nov 5 13:37:41 blade414 0x43d7c2: /usr/sbin/varnishd(MGT_Run+0x132)
[0x43d7c2]

}}}


{{{
USER PID %CPU %MEM VSZ RSS TTY STAT START TIME COMMAND
root 28818 0.0 0.0 31570672 1296 ? Ss Oct31 0:00
/usr/sbin/varnishd
}}}

{{{
-P /var/run/varnish.pid
-f /etc/varnish/default.vcl
-p lru_interval 20
-p cli_timeout 60
-h classic,500009
-s malloc,30G
-s persistent,/var/lib/varnish/store/silo,30G
-p thread_pool_min 500
-p thread_pool_max 4000
-p thread_pools 4
-p thread_pool_add_delay 2
-p session_linger 100
-p send_timeout 120
-p sess_timeout 10
-p listen_depth 4096
-p nuke_limit 50
}}}

--
Ticket URL: <https://www.varnish-cache.org/trac/ticket/1051>
Varnish <https://varnish-cache.org/>
The Varnish HTTP Accelerator

_______________________________________________
varnish-bugs mailing list
varnish-bugs [at] varnish-cache
https://www.varnish-cache.org/lists/mailman/listinfo/varnish-bugs


varnish-bugs at varnish-cache

Nov 5, 2011, 6:58 AM

Post #2 of 8 (148 views)
Permalink
Re: #1051: child process died [In reply to]

#1051: child process died
---------------------+------------------------------------------------------
Reporter: sreniaw | Type: defect
Status: new | Priority: normal
Milestone: | Component: varnishd
Version: 3.0.2 | Severity: normal
Keywords: |
---------------------+------------------------------------------------------

Comment(by sreniaw):

I am not able to start varnish with persistent cache.
{{{
# df -h /var/lib/varnish/store/
Filesystem Size Used Avail Use% Mounted on
/dev/mapper/Varnish
40G 28G 9.7G 75% /var/lib/varnish/store
# ps
root 26146 0.0 0.0 31569644 1208 ? Ss 14:52 0:00
/usr/sbin/varnishd
}}}

After disabling the option "-s" varnish is starting.

{{{
root 23326 0.0 0.0 112368 1240 ? Ss 14:50 0:00
/usr/sbin/varnishd
varnish 23327 0.2 0.3 20746912 106348 ? Sl 14:50 0:00 \_
/usr/sbin/varnishd
}}}

{{{
Nov 5 14:52:31 blade414 CLI communication error (hdr)
Nov 5 14:52:31 blade414 Condition(sg1->p.offset != sg->p.offset) not
true.
Nov 5 14:52:31 blade414 thread = (cache-main)
Nov 5 14:52:31 blade414 ident =
Linux,2.6.18-238.el5,x86_64,-smalloc,-spersistent,-smalloc,-hclassic,no_waiter
Nov 5 14:52:31 blade414 Backtrace:
Nov 5 14:52:31 blade414 0x42c7a6: /usr/sbin/varnishd [0x42c7a6]
Nov 5 14:52:31 blade414 0x4472f5: /usr/sbin/varnishd [0x4472f5]
Nov 5 14:52:31 blade414 0x4474bb: /usr/sbin/varnishd [0x4474bb]
Nov 5 14:52:31 blade414 0x444d57: /usr/sbin/varnishd(STV_open+0x27)
[0x444d57]
Nov 5 14:52:31 blade414 0x42b525: /usr/sbin/varnishd(child_main+0xc5)
[0x42b525]
Nov 5 14:52:31 blade414 0x43d5ec: /usr/sbin/varnishd [0x43d5ec]
Nov 5 14:52:31 blade414 0x43d852: /usr/sbin/varnishd(MGT_Run+0x1c2)
[0x43d852]
Nov 5 14:52:31 blade414 0x44cacb: /usr/sbin/varnishd(main+0xd1b)
[0x44cacb]
Nov 5 14:52:31 blade414 0x38bda1d994:
/lib64/libc.so.6(__libc_start_main+0xf4) [0x38bda1d994]
Nov 5 14:52:31 blade414 0x40ba79: /usr/sbin/varnishd(VCLS_func_help+0x81)
[0x40ba79]
Nov 5 14:52:33 blade414 CLI communication error (hdr)
Nov 5 14:52:33 blade414 Condition(sg1->p.offset != sg->p.offset) not
true.
Nov 5 14:52:33 blade414 thread = (cache-main)
Nov 5 14:52:33 blade414 ident =
Linux,2.6.18-238.el5,x86_64,-smalloc,-spersistent,-smalloc,-hclassic,no_waiter
Nov 5 14:52:33 blade414 Backtrace:
Nov 5 14:52:33 blade414 0x42c7a6: /usr/sbin/varnishd [0x42c7a6]
Nov 5 14:52:33 blade414 0x4472f5: /usr/sbin/varnishd [0x4472f5]
Nov 5 14:52:33 blade414 0x4474bb: /usr/sbin/varnishd [0x4474bb]
Nov 5 14:52:33 blade414 0x444d57: /usr/sbin/varnishd(STV_open+0x27)
[0x444d57]
Nov 5 14:52:33 blade414 0x42b525: /usr/sbin/varnishd(child_main+0xc5)
[0x42b525]
Nov 5 14:52:33 blade414 0x43d5ec: /usr/sbin/varnishd [0x43d5ec]
Nov 5 14:52:33 blade414 0x43d852: /usr/sbin/varnishd(MGT_Run+0x1c2)
[0x43d852]
Nov 5 14:52:33 blade414 0x44cacb: /usr/sbin/varnishd(main+0xd1b)
[0x44cacb]
Nov 5 14:52:33 blade414 0x38bda1d994:
/lib64/libc.so.6(__libc_start_main+0xf4) [0x38bda1d994]
Nov 5 14:52:33 blade414 0x40ba79: /usr/sbin/varnishd(VCLS_func_help+0x81)
[0x40ba79]
}}}

--
Ticket URL: <https://www.varnish-cache.org/trac/ticket/1051#comment:1>
Varnish <https://varnish-cache.org/>
The Varnish HTTP Accelerator

_______________________________________________
varnish-bugs mailing list
varnish-bugs [at] varnish-cache
https://www.varnish-cache.org/lists/mailman/listinfo/varnish-bugs


varnish-bugs at varnish-cache

Nov 7, 2011, 3:13 AM

Post #3 of 8 (150 views)
Permalink
Re: #1051: child process died [In reply to]

#1051: child process died
----------------------+-----------------------------------------------------
Reporter: sreniaw | Type: defect
Status: closed | Priority: normal
Milestone: | Component: varnishd
Version: 3.0.2 | Severity: normal
Resolution: fixed | Keywords:
----------------------+-----------------------------------------------------
Changes (by phk):

* status: new => closed
* resolution: => fixed


Comment:

Your -spersistent silo has been corrupted, and you should remove the file
entirely.

Please tell me if it happens again.

--
Ticket URL: <https://www.varnish-cache.org/trac/ticket/1051#comment:2>
Varnish <https://varnish-cache.org/>
The Varnish HTTP Accelerator

_______________________________________________
varnish-bugs mailing list
varnish-bugs [at] varnish-cache
https://www.varnish-cache.org/lists/mailman/listinfo/varnish-bugs


varnish-bugs at varnish-cache

Jan 17, 2012, 7:16 AM

Post #4 of 8 (108 views)
Permalink
Re: #1051: child process died [In reply to]

#1051: child process died
-----------------------+----------------------------------------------------
Reporter: sreniaw | Type: defect
Status: reopened | Priority: normal
Milestone: | Component: varnishd
Version: 3.0.2 | Severity: normal
Resolution: | Keywords:
-----------------------+----------------------------------------------------
Changes (by sreniaw2):

* status: closed => reopened
* resolution: fixed =>


Comment:

I am having problem with persistent cache.
Silo file is created using dd command with initial size of 35GBs.
After two weeks varnish is starting crashing.

I am using varnish-3.0.2-1.el5 on Centos Linux 5 x86_64 with 32 GB of
physical memory and 33 GB of swap.

dm-7 is only for silo file
{{{
# iostat -x 1
Device: rrqm/s wrqm/s r/s w/s rsec/s wsec/s avgrq-sz
avgqu-sz await svctm %util
dm-7 0.00 0.00 1.29 163.92 111.77 1311.38 8.61
0.60 3.64 1.29 21.30
dm-7 0.00 0.00 0.00 662.00 0.00 5296.00 8.00
113.08 171.04 1.51 100.10
dm-7 0.00 0.00 2.00 646.00 256.00 5168.00 8.37
116.20 177.84 1.54 100.10
dm-7 0.00 0.00 0.00 647.52 0.00 5180.20 8.00
113.79 182.72 1.53 99.11
dm-7 0.00 0.00 4.00 640.00 312.00 5120.00 8.43
94.62 131.55 1.55 100.10
dm-7 0.00 0.00 2.00 620.00 216.00 4960.00 8.32
118.29 202.90 1.61 100.20
dm-7 0.00 0.00 4.00 645.00 416.00 5160.00 8.59
114.23 179.15 1.54 100.10
dm-7 0.00 0.00 0.00 611.00 0.00 4888.00 8.00
112.30 164.37 1.64 100.10
dm-7 0.00 0.00 3.00 581.00 424.00 4648.00 8.68
113.39 209.71 1.71 100.00
dm-7 0.00 0.00 4.00 594.00 512.00 4752.00 8.80
91.58 168.89 1.67 100.10
dm-7 0.00 0.00 2.00 708.00 112.00 5664.00 8.14
115.57 158.32 1.41 100.10
dm-7 0.00 0.00 4.00 658.00 376.00 5264.00 8.52
115.12 166.39 1.51 100.10
dm-7 0.00 0.00 2.00 527.00 256.00 4216.00 8.45
114.82 213.90 1.89 100.10
dm-7 0.00 0.00 0.00 680.00 0.00 5440.00 8.00
115.64 171.43 1.47 100.20
dm-7 0.00 0.00 19.80 599.01 633.66 4792.08 8.77
101.46 172.11 1.60 99.21
dm-7 0.00 0.00 1.00 664.00 128.00 5312.00 8.18
115.89 167.74 1.51 100.20
dm-7 0.00 0.00 5.00 622.00 360.00 4976.00 8.51
114.20 172.12 1.60 100.10
dm-7 0.00 0.00 3.00 582.00 224.00 4656.00 8.34
112.75 205.01 1.71 100.10
dm-7 0.00 0.00 0.00 667.00 0.00 5336.00 8.00
110.22 164.68 1.50 100.20
dm-7 0.00 0.00 1.00 669.00 128.00 5352.00 8.18
103.50 155.61 1.50 100.20
dm-7 0.00 0.00 3.00 596.00 240.00 4768.00 8.36
113.84 184.50 1.67 100.10
dm-7 0.00 0.00 10.89 564.36 807.92 4514.85 9.25
114.11 203.60 1.72 99.21
dm-7 0.00 0.00 4.00 617.00 264.00 4936.00 8.37
116.19 184.58 1.61 100.10
dm-7 0.00 0.00 3.00 607.00 224.00 4856.00 8.33
112.08 186.60 1.64 100.00
dm-7 0.00 0.00 0.00 698.00 0.00 5584.00 8.00
113.24 145.28 1.44 100.20
dm-7 0.00 0.00 0.00 6062.00 0.00 48496.00 8.00
1007.00 106.46 0.17 100.10
dm-7 0.00 0.00 0.00 750.00 0.00 6000.00 8.00
268.74 834.84 1.34 100.20
dm-7 0.00 0.00 0.00 656.00 0.00 5248.00 8.00
115.95 209.27 1.52 100.00
dm-7 0.00 0.00 0.00 730.00 0.00 5840.00 8.00
152.54 187.88 1.37 100.20
dm-7 0.00 0.00 15.84 55.45 2146.53 443.56 36.33
21.85 652.62 6.14 43.76
dm-7 0.00 0.00 4.00 0.00 512.00 0.00 128.00
0.03 6.75 6.75 2.70
dm-7 0.00 0.00 0.00 0.00 0.00 0.00 0.00
0.00 0.00 0.00 0.00
dm-7 0.00 0.00 3.00 0.00 288.00 0.00 96.00
0.00 1.33 1.33 0.40
dm-7 0.00 0.00 3.00 456.00 384.00 3648.00 8.78
38.76 64.82 0.66 30.40
dm-7 0.00 0.00 0.00 991.00 0.00 7928.00 8.00
119.20 120.89 1.01 100.20
dm-7 0.00 0.00 2.00 945.00 112.00 7560.00 8.10
114.86 114.68 1.06 100.20
dm-7 0.00 0.00 4.00 834.00 216.00 6672.00 8.22
118.92 138.72 1.19 100.10
dm-7 0.00 0.00 2.97 743.56 205.94 5948.51 8.24
117.64 156.17 1.33 99.21
dm-7 0.00 0.00 11.00 681.00 592.00 5448.00 8.73
117.81 176.87 1.45 100.00
}}}

{{{
# ps axu | grep varnishd
root 6462 0.0 0.0 34716380 1392 ? Ss Jan02 0:00
/usr/sbin/varnishd -P /var/run/varnish.pid -a :80 -T localhost:6082 -f
/etc/varnish/default.vcl -u varnish
+-g varnish -S /etc/varnish/secret -p lru_interval 20 -p cli_timeout 60 -h
classic,500009 -s malloc,30G -s
persistent,/var/lib/varnish/store/silo,35433480192 -p
+thread_pool_min 500 -p thread_pool_max 4000 -p thread_pools 4 -p
thread_pool_add_delay 2 -p session_linger 100 -p send_timeout 120 -p
sess_timeout 10 -p listen_depth 4096
+-p nuke_limit 50
varnish 29817 25.9 0.0 0 0 ? Zl 22:29 4:48
[varnishd] <defunct>
}}}

{{{
# ps axu | grep varnishd
varnish 1870 25.4 2.4 55950024 808356 ? Sl 22:48 0:09
/usr/sbin/varnishd -P /var/run/varnish.pid -a :80 -T localhost:6082 -f
/etc/varnish/default.vcl -u varnish
+-g varnish -S /etc/varnish/secret -p lru_interval 20 -p cli_timeout 60 -h
classic,500009 -s malloc,30G -s
persistent,/var/lib/varnish/store/silo,35433480192 -p
+thread_pool_min 500 -p thread_pool_max 4000 -p thread_pools 4 -p
thread_pool_add_delay 2 -p session_linger 100 -p send_timeout 120 -p
sess_timeout 10 -p listen_depth 4096
+-p nuke_limit 50
root 3969 0.0 0.0 61172 728 pts/3 R+ 22:48 0:00 grep
varnishd
root 6462 0.0 0.0 34716380 1400 ? Ss Jan02 0:00
/usr/sbin/varnishd -P /var/run/varnish.pid -a :80 -T localhost:6082 -f
/etc/varnish/default.vcl -u varnish
+-g varnish -S /etc/varnish/secret -p lru_interval 20 -p cli_timeout 60 -h
classic,500009 -s malloc,30G -s
persistent,/var/lib/varnish/store/silo,35433480192 -p
+thread_pool_min 500 -p thread_pool_max 4000 -p thread_pools 4 -p
thread_pool_add_delay 2 -p session_linger 100 -p send_timeout 120 -p
sess_timeout 10 -p listen_depth 4096
+-p nuke_limit 50
}}}

Next day:
{{{
# ps axu | grep varnishd
root 6462 0.0 0.0 34716380 1524 ? Ss Jan02 0:00
/usr/sbin/varnishd -P /var/run/varnish.pid -a :80 -T localhost:6082 -f
/etc/varnish/default.vcl -u varnish -g varnish -S /etc/varnish/secret -p
lru_interval 20 -p cli_timeout 60 -h classic,500009 -s malloc,30G -s
persistent,/var/lib/varnish/store/silo,35433480192 -p thread_pool_min 500
-p thread_pool_max 4000 -p thread_pools 4 -p thread_pool_add_delay 2 -p
session_linger 100 -p send_timeout 120 -p sess_timeout 10 -p listen_depth
4096 -p nuke_limit 50
varnish 14624 11.4 17.9 56139464 5921276 ? Sl 15:14 4:29
/usr/sbin/varnishd -P /var/run/varnish.pid -a :80 -T localhost:6082 -f
/etc/varnish/default.vcl -u varnish -g varnish -S /etc/varnish/secret -p
lru_interval 20 -p cli_timeout 60 -h classic,500009 -s malloc,30G -s
persistent,/var/lib/varnish/store/silo,35433480192 -p thread_pool_min 500
-p thread_pool_max 4000 -p thread_pools 4 -p thread_pool_add_delay 2 -p
session_linger 100 -p send_timeout 120 -p sess_timeout 10 -p listen_depth
4096 -p nuke_limit 50
}}}

{{{
DAEMON_OPTS="-a :80 \
-T localhost:6082 \
-f /etc/varnish/default.vcl \
-u varnish -g varnish \
-S /etc/varnish/secret \
-p lru_interval=20 \
-p cli_timeout=60 \
-h classic,500009 \
-s malloc,30G -s
persistent,/var/lib/varnish/store/silo,35433480192 \
-p thread_pool_min=500 \
-p thread_pool_max=4000 \
-p thread_pools=4 \
-p thread_pool_add_delay=2 \
-p session_linger=100 \
-p send_timeout=120 \
-p sess_timeout=10 \
-p listen_depth=4096 \
-p nuke_limit=50"
}}}

Any clues ?

--
Ticket URL: <https://www.varnish-cache.org/trac/ticket/1051#comment:3>
Varnish <https://varnish-cache.org/>
The Varnish HTTP Accelerator

_______________________________________________
varnish-bugs mailing list
varnish-bugs [at] varnish-cache
https://www.varnish-cache.org/lists/mailman/listinfo/varnish-bugs


varnish-bugs at varnish-cache

Jan 17, 2012, 7:36 AM

Post #5 of 8 (108 views)
Permalink
Re: #1051: child process died [In reply to]

#1051: child process died
-----------------------+----------------------------------------------------
Reporter: sreniaw | Type: defect
Status: reopened | Priority: normal
Milestone: | Component: varnishd
Version: 3.0.2 | Severity: normal
Resolution: | Keywords:
-----------------------+----------------------------------------------------

Comment(by sreniaw2):

Logs provides clues:
Jan 17 14:14:31 varnishd[6462]: Child (14624) said Dropped 10 segments to
make free_reserve
Jan 17 14:14:32 varnishd[6462]: Child (14624) said Silo completely loaded
Jan 17 15:03:00 varnishd[6462]: Child (14624) said Out of space in
persistent silo
Jan 17 15:03:00 varnishd[6462]: Child (14624) said Committing suicide,
restart will make space


{{{
# du -sh /var/lib/varnish/store/silo
34G /var/lib/varnish/store/silo
}}}

{{{
# df -h /var/lib/varnish/store
Filesystem Size Used Avail Use% Mounted on
/dev/mapper/SysStor-Varnish
36G 34G 442M 99% /var/lib/varnish/store
}}}

--
Ticket URL: <https://www.varnish-cache.org/trac/ticket/1051#comment:4>
Varnish <https://varnish-cache.org/>
The Varnish HTTP Accelerator

_______________________________________________
varnish-bugs mailing list
varnish-bugs [at] varnish-cache
https://www.varnish-cache.org/lists/mailman/listinfo/varnish-bugs


varnish-bugs at varnish-cache

Jan 17, 2012, 7:37 AM

Post #6 of 8 (112 views)
Permalink
Re: #1051: child process died [In reply to]

#1051: child process died
-----------------------+----------------------------------------------------
Reporter: sreniaw | Type: defect
Status: reopened | Priority: normal
Milestone: | Component: varnishd
Version: 3.0.2 | Severity: normal
Resolution: | Keywords:
-----------------------+----------------------------------------------------

Comment(by sreniaw2):

{{{
Jan 17 14:14:31 varnishd[6462]: Child (14624) said Dropped 10 segments to
make free_reserve
Jan 17 14:14:32 varnishd[6462]: Child (14624) said Silo completely loaded
Jan 17 15:03:00 varnishd[6462]: Child (14624) said Out of space in
persistent silo
Jan 17 15:03:00 varnishd[6462]: Child (14624) said Committing suicide,
restart will make space
}}}

--
Ticket URL: <https://www.varnish-cache.org/trac/ticket/1051#comment:5>
Varnish <https://varnish-cache.org/>
The Varnish HTTP Accelerator

_______________________________________________
varnish-bugs mailing list
varnish-bugs [at] varnish-cache
https://www.varnish-cache.org/lists/mailman/listinfo/varnish-bugs


varnish-bugs at varnish-cache

Jan 23, 2012, 5:28 AM

Post #7 of 8 (109 views)
Permalink
Re: #1051: child process died [In reply to]

#1051: child process died
----------------------+-----------------------------------------------------
Reporter: sreniaw | Type: defect
Status: closed | Priority: normal
Milestone: | Component: varnishd
Version: 3.0.2 | Severity: normal
Resolution: fixed | Keywords:
----------------------+-----------------------------------------------------
Changes (by kristian):

* status: reopened => closed
* resolution: => fixed


Comment:

@sreniaw2:

What you are experiencing is running out of space, and persistent storage
currently has no code to handle it, therefor it restarts.

The fix PHK committed fixed corruption, which made restarting problematic,
but does not solve the lack of LRU, which is distinct and a feature
request.

The only remedy for this is to make sure you do not run out of space.
Either by caching content for a shorter duration or by increasing the
available space.

As this is a known limitation of persistent storage, and the correct code
to fix it is a feature request, I'm closing this. This does not indicate
we're not planning to fix it, just that we do not track feature requests
in trac.

I'm closing this as "fixed" because the corruption and origional assert
should have been fixed. The LRU-thing is not.

--
Ticket URL: <https://www.varnish-cache.org/trac/ticket/1051#comment:6>
Varnish <https://varnish-cache.org/>
The Varnish HTTP Accelerator

_______________________________________________
varnish-bugs mailing list
varnish-bugs [at] varnish-cache
https://www.varnish-cache.org/lists/mailman/listinfo/varnish-bugs


varnish-bugs at varnish-cache

Mar 15, 2012, 7:46 AM

Post #8 of 8 (61 views)
Permalink
Re: #1051: child process died [In reply to]

#1051: child process died
-----------------------+----------------------------------------------------
Reporter: sreniaw | Type: defect
Status: reopened | Priority: normal
Milestone: | Component: varnishd
Version: 3.0.2 | Severity: normal
Resolution: | Keywords:
-----------------------+----------------------------------------------------
Changes (by sreniaw2):

* status: closed => reopened
* resolution: fixed =>


Comment:

/var/lib/pgsql/${postgres_version}/data/postgresql.conf

Replying to [comment:2 phk]:
> Your -spersistent silo has been corrupted, and you should remove the
file entirely.
>
> Please tell me if it happens again.

We have experienced again same crash.
Child process died and it wasn't able to recover.
The problem was solved by deleting silo.

{{{
Mar 13 21:17:14 CLI communication error (hdr)
Mar 13 21:17:14 Condition(sg1->p.offset != sg->p.offset) not true.
Mar 13 21:17:14 thread = (cache-main)
Mar 13 21:17:14 ident =
Linux,2.6.18-274.18.1.el5,x86_64,-smalloc,-spersistent,-smalloc,-hclassic,no_waiter
Mar 13 21:17:14 Backtrace:
Mar 13 21:17:14 0x42c7a6: /usr/sbin/varnishd [0x42c7a6]
Mar 13 21:17:14 0x4472f5: /usr/sbin/varnishd [0x4472f5]
Mar 13 21:17:14 0x4474bb: /usr/sbin/varnishd [0x4474bb]
Mar 13 21:17:14 0x444d57: /usr/sbin/varnishd(STV_open+0x27) [0x444d57]
Mar 13 21:17:14 0x42b525: /usr/sbin/varnishd(child_main+0xc5) [0x42b525]
Mar 13 21:17:14 0x43d5ec: /usr/sbin/varnishd [0x43d5ec]
Mar 13 21:17:14 0x43d852: /usr/sbin/varnishd(MGT_Run+0x1c2) [0x43d852]
Mar 13 21:17:14 0x44cacb: /usr/sbin/varnishd(main+0xd1b) [0x44cacb]
Mar 13 21:17:14 0x3411c1d994: /lib64/libc.so.6(__libc_start_main+0xf4)
[0x3411c1d994]
Mar 13 21:17:14 0x40ba79: /usr/sbin/varnishd(VCLS_func_help+0x81)
[0x40ba79]
}}}



{{{
# ps axu | grep varnishd
root 21704 0.0 0.0 63026924 1216 ? Ss 21:17 0:00
/usr/sbin/varnishd -P
/var/run/varnish.pid -a :80 -T localhost:6082 -f /etc/varnish/default.vcl
-u varnish -g
varnish -S /etc/varnish/secret -p lru_interval 20 -p cli_timeout 60 -h
classic,500009 -s
malloc,30G -s persistent,/var/lib/varnish/store/silo,64424509440 -p
thread_pool_min 500 -p
thread_pool_max 4000 -p thread_pools 4 -p thread_pool_add_delay 2 -p
session_linger 100 -p
send_timeout 120 -p sess_timeout 10 -p listen_depth 4096 -p nuke_limit 50
root 21786 0.0 0.0 61140 704 pts/5 S+ 21:17 0:00 grep
varnishd

}}}

--
Ticket URL: <https://www.varnish-cache.org/trac/ticket/1051#comment:7>
Varnish <https://varnish-cache.org/>
The Varnish HTTP Accelerator

_______________________________________________
varnish-bugs mailing list
varnish-bugs [at] varnish-cache
https://www.varnish-cache.org/lists/mailman/listinfo/varnish-bugs

Varnish bugs RSS feed   Index | Next | Previous | View Threaded
 
 


Interested in having your list archived? Contact Gossamer Threads
 
  Web Applications & Managed Hosting Powered by Gossamer Threads Inc.