Re: memory leak in Init 10.3-STABLE .svn_revision 304147 ?
Hi, Reference: > From: Eugene Grosbein <eu...@grosbein.net> > Date: Sat, 3 Mar 2018 23:48:42 +0700 Eugene Grosbein wrote: > 03.03.2018 23:33, Julian H. Stacey wrote: > > > Eugene Grosbein wrote: > >> 03.03.2018 23:11, Julian H. Stacey wrote: > >> > >>> Oops! I got Subject & body wrong s/inetd/init/ ! > >>> > >>> Was: memory leak in inetd 10.3-STABLE .svn_revision 304147 ? > >>> Should be: memory leak in init 10.3-STABLE .svn_revision 304147 ? > >> > >> There is no evidence that /sbin/init has memory leak (nor inetd), what > >> makes you think so? > > > > I got back about 500M of wap after kill -HUP 1 > > There may be plenty of reasons for that apart from init's "memory leak" > and init does not even show in the top. Please share your /etc/ttys file. console noneunknown off secure ttyv0 "/usr/libexec/getty Pc" xterm on secure ttyv1 "/usr/libexec/getty Pc" xterm on secure ttyv2 "/usr/libexec/getty Pc" xterm on secure ttyv3 "/usr/libexec/getty Pc" xterm off secure ttyv4 "/usr/libexec/getty Pc" xterm off secure ttyv5 "/usr/libexec/getty Pc" xterm off secure ttyv6 "/usr/libexec/getty Pc" xterm off secure ttyv7 "/usr/libexec/getty Pc" xterm off secure ttyv8 "/usr/local/bin/xdm -nodaemon" xterm off secure ttyu0 "/usr/libexec/getty 3wire" vt100 onifconsole secure ttyu1 "/usr/libexec/getty 3wire" vt100 onifconsole secure ttyu2 "/usr/libexec/getty 3wire" vt100 onifconsole secure ttyu3 "/usr/libexec/getty 3wire" vt100 onifconsole secure dcons "/usr/libexec/getty std.9600" vt100 off secure Cheers, Julian -- Julian Stacey, Computer Consultant, Systems Engineer, BSD Linux Unix, Munich Brexit Referendum stole 3,700,000 votes, inc. 700,000 from British in EU. UK Govt. lied it's "democratic" in Article 50 letter to EU paragraph 3. Petition for votes: http://berklix.eu/queen/ ___ freebsd-stable@freebsd.org mailing list https://lists.freebsd.org/mailman/listinfo/freebsd-stable To unsubscribe, send any mail to "freebsd-stable-unsubscr...@freebsd.org"
Re: memory leak in Init 10.3-STABLE .svn_revision 304147 ?
03.03.2018 23:33, Julian H. Stacey wrote: > Eugene Grosbein wrote: >> 03.03.2018 23:11, Julian H. Stacey wrote: >> >>> Oops! I got Subject & body wrong s/inetd/init/ ! >>> >>> Was:memory leak in inetd 10.3-STABLE .svn_revision 304147 ? >>> Should be: memory leak in init 10.3-STABLE .svn_revision 304147 ? >> >> There is no evidence that /sbin/init has memory leak (nor inetd), what makes >> you think so? > > I got back about 500M of wap after kill -HUP 1 There may be plenty of reasons for that apart from init's "memory leak" and init does not even show in the top. Please share your /etc/ttys file. ___ freebsd-stable@freebsd.org mailing list https://lists.freebsd.org/mailman/listinfo/freebsd-stable To unsubscribe, send any mail to "freebsd-stable-unsubscr...@freebsd.org"
Re: memory leak in Init 10.3-STABLE .svn_revision 304147 ?
Eugene Grosbein wrote: > 03.03.2018 23:11, Julian H. Stacey wrote: > > > Oops! I got Subject & body wrong s/inetd/init/ ! > > > > Was:memory leak in inetd 10.3-STABLE .svn_revision 304147 ? > > Should be: memory leak in init 10.3-STABLE .svn_revision 304147 ? > > There is no evidence that /sbin/init has memory leak (nor inetd), what makes > you think so? I got back about 500M of wap after kill -HUP 1 > Pleas show output of "top -ores -d1" command. last pid: 57791; load averages: 0.39, 0.52, 0.48 up 547+03:40:39 17:31:32 41 processes: 2 running, 39 sleeping CPU: % user, % nice, % system, % interrupt, % idle Mem: 32M Active, 1160M Inact, 388M Wired, 108K Cache, 212M Buf, 391M Free Swap: 614M Total, 79M Used, 535M Free, 12% Inuse PID USERNAMETHR PRI NICE SIZERES STATETIMEWCPU COMMAND 47677 bind 4 200 74956K 28088K kqread 231:10 0.00% named 35591 mailman 1 200 113M 18968K select 47:57 0.00% python2. 35590 mailman 1 200 112M 16120K select 50:09 0.00% python2. 35588 mailman 1 200 112M 15588K select 51:08 0.00% python2. 35592 mailman 1 200 108M 14928K select 51:53 0.00% python2. 579 root 1 200 21900K 13812K select 867:23 0.00% ntpd 35593 mailman 1 200 107M 13564K select 50:30 0.00% python2. 35589 mailman 1 200 108M 13432K select 54:56 0.00% python2. 96114 root 1 200 101M 11388K select 8:20 0.00% vmtoolsd 35587 mailman 1 200 106M 10832K wait 0:01 0.00% python2. 57188 www 1 200 147M 9708K lockf0:01 0.00% httpd 57337 www 1 200 147M 9692K select 0:01 0.00% httpd 57335 www 1 200 147M 9672K lockf0:01 0.00% httpd 57616 www 1 200 147M 9660K select 0:00 0.00% httpd 57365 www 1 200 147M 9652K select 0:00 0.00% httpd 57613 www 1 200 147M 9648K kqread 0:00 0.00% httpd 78720 root 1 200 147M 9512K select 0:40 0.00% httpd Cheers, Julian -- Julian Stacey, Computer Consultant, Systems Engineer, BSD Linux Unix, Munich Brexit Referendum stole 3,700,000 votes, inc. 700,000 from British in EU. UK Govt. lied it's "democratic" in Article 50 letter to EU paragraph 3. Petition for votes: http://berklix.eu/queen/ ___ freebsd-stable@freebsd.org mailing list https://lists.freebsd.org/mailman/listinfo/freebsd-stable To unsubscribe, send any mail to "freebsd-stable-unsubscr...@freebsd.org"
Re: memory leak in Init 10.3-STABLE .svn_revision 304147 ?
03.03.2018 23:11, Julian H. Stacey wrote: > Oops! I got Subject & body wrong s/inetd/init/ ! > > Was:memory leak in inetd 10.3-STABLE .svn_revision 304147 ? > Should be: memory leak in init 10.3-STABLE .svn_revision 304147 ? There is no evidence that /sbin/init has memory leak (nor inetd), what makes you think so? Pleas show output of "top -ores -d1" command. ___ freebsd-stable@freebsd.org mailing list https://lists.freebsd.org/mailman/listinfo/freebsd-stable To unsubscribe, send any mail to "freebsd-stable-unsubscr...@freebsd.org"
memory leak in Init 10.3-STABLE .svn_revision 304147 ?
Oops! I got Subject & body wrong s/inetd/init/ ! Was:memory leak in inetd 10.3-STABLE .svn_revision 304147 ? Should be: memory leak in init 10.3-STABLE .svn_revision 304147 ? Hi sta...@freebsd.org I think I have found a memory leak with init on 10.3-STABLE (server runs in a VM); server ran out of swap. top Mem: 34M Active, 791M Inact, 387M Wired, 200K Cache, 212M Buf, 759M Free Swap: 614M Total, 605M Used, 8744K Free, 98% Inuse uptime 4:47PM up 544 days, 2:56, 2 users, load averages: 0.26, 0.30, 0.32 uname -a FreeBSD slim.domain 10.3-STABLE FreeBSD 10.3-STABLE #0: Tue Aug 16 18:09:22 CEST 2016 jhs@slim.domain:/usr/obj/usr/src/sys/GENERIC amd64 First I just commented out a few trivial getty as a start, ran kill -HUP 1 and was surprised to get loads of swap back: swapinfo Device 1K-blocks UsedAvail Capacity /dev/da0p3 62873681056 54768013% ( Before I ran `kill -9 541` after seeing this: PID USERNAMETHR PRI NICE SIZERES STATETIMEWCPU COMMAND 541 root 2 520 769M 103M select 28.4H 0.00% vmtoolsd but it didn't gain me swap space, hence next the `kill -HUP 1` which I had first avoided, as server is remote. Restarting vmtoolsd leaves swapinfo unchanged. ) top CPU: 0.8% user, 0.0% nice, 3.1% system, 0.0% interrupt, 96.1% idle Mem: 32M Active, 711M Inact, 383M Wired, 176K Cache, 216M Buf, 846M Free Swap: 614M Total, 79M Used, 535M Free, 12% Inuse /var/log/dmesg.today 858 x swap_pager_getswapspace In case someone asks me to de specific debugging, that server has: cd /usr/src cat .svn_revision 304147 cat .ctm_status src-10 1304 Though I'm wary of a reboot with a debug kernel in case it hangs, as it's a remote live server. (So I guess an ideal world would have someone say there used to be a leak but it was fixed long since.) PS Below there's an intriguing "Swap: 32G", presumably coming from hypervisor outside my server. - BEFORE top last pid: 91299; load averages: 0.62, 0.68, 0.49 up 543+22:18:42 12:09:35 47 processes: 1 running, 46 sleeping Mem: 34M Active, 791M Inact, 387M Wired, 200K Cache, 212M Buf, 759M Free Swap: 614M Total, 605M Used, 8744K Free, 98% Inuse PID USERNAMETHR PRI NICE SIZERES STATETIMEWCPU COMMAND 541 root 2 520 769M 103M select 28.4H 0.00% vmtoolsd 579 root 1 200 21900K 13812K select 862:04 0.00% ntpd 47677 bind 4 200 74956K 27996K kqread 227:10 0.00% named 411 root 1 200 14524K 1708K select 124:23 0.00% syslogd 313 root 1 200 13656K 536K select 121:22 0.00% devd 63839 root 1 200 51512K 4852K select 61:12 0.00% sendmail 35589 mailman 1 200 108M 13168K select 53:29 0.00% python2. 35592 mailman 1 200 108M 14852K select 50:27 0.00% python2. 35588 mailman 1 200 112M 15384K select 49:43 0.00% python2. 35593 mailman 1 200 107M 13344K select 49:10 0.00% python2. 35590 mailman 1 200 112M 15272K select 48:49 0.00% python2. 35591 mailman 1 200 113M 18564K select 46:37 0.00% python2. 651 root 1 200 61316K 4116K select 17:21 0.00% sshd 676 root 1 200 16624K 2228K nanslp 11:02 0.00% cron 63843 smmsp 1 200 26480K 4048K pause0:30 0.00% sendmail 35594 mailman 1 200 106M 5932K select 0:27 0.00% python2. 78594 root 1 200 16832K 2576K kqread 0:15 0.00% dovecot 78720 root 1 200 147M 9512K select 0:09 0.00% httpd - IMMEDIATELY BEFORE last pid: 12063; load averages: 0.02, 0.06, 0.07 up 360+17:30:09 12:05:23 23 processes: 1 running, 22 sleeping CPU: 0.0% user, 0.0% nice, 0.1% system, 0.0% interrupt, 99.9% idle Mem: 242M Active, 18G Inact, 12G Wired, 161M Cache, 240M Free ARC: 8578M Total, 3785M MFU, 809M MRU, 59K Anon, 227M Header, 3758M Other Swap: 32G Total, 200M Used, 32G Free PID USERNAME THR PRI NICE SIZERES STATE C TIMEWCPU COMMAN 97336 bind11 200 109M 30084K kqread 4 57:01 0.00% named 1544 root 1 200 14536K 656K select 7 10:19 0.00% syslog 47145 root 1 200 41596K 3008K select 2 7:42 0.00% sendma 22566 root 1 200 16836K 1324K kqread 6 7:00 0.00% doveco 22567 root 1 200 21264K 1176K kqread 6 2:28 0.00% doveco 1704 root 1 200 34636K 744K select 2 1:31 0.00% sshd 6181 root 1 200 149M 7192K select 4 1:00 0.00% httpd 1764 root 1 200 18752K 708K select 2