Re: Workers dying with "out of memory in new"
"Johanna Amann" <[email protected]> Fri, 18 Oct 2019 16:38:24 -0700
| Newsgroups | gmane.comp.security.detection.bro |
|---|---|
| Message-ID | <[email protected]> |
Hi, both of you are running rather old versions of Zeek. Both 2.5.5 and 2.6.1 have a number of issues. One of the issues that was fixed could be the cause for crashes. A bug could result in Zeek requesting huge allocations that cannot be fulfilled by the operating system; see e.g. https://github.com/zeek/zeek/issues/245 for more details. This specific issue was fixed on 2.6.3. So - upgrading to 2.6.4 (or even better - 3.0.0) might fix those problems for you. Besides that - both 2.5.5 and 2.6.1 have several vulnerabilities - and you really really really should upgrade them :). Johanna On 18 Oct 2019, at 8:26, Munroe Sollog wrote: > For additional reference: > > Linux snout 4.9.0-9-amd64 #1 SMP Debian 4.9.168-1+deb9u5 (2019-08-11) > x86_64 GNU/Linux > > on 10-11 I patched libssl,and libc > on 10-17 I upgraded sudo (about 30 mins after the first worker > crashed) > > [Bro] Crash report from worker-1-12 email received at 16:00 > > Log output from dpkg for reference: > > # less /var/log/dpkg.log |grep "status installed" > > 2019-10-11 14:59:23 status installed telegraf:amd64 1.12.3-1 > > 2019-10-11 14:59:23 status installed libssl1.0.2:amd64 1.0.2t-1~deb9u1 > > 2019-10-11 14:59:23 status installed libc-bin:amd64 2.24-11+deb9u4 > > 2019-10-11 14:59:23 status installed libssl1.1:amd64 1.1.0l-1~deb9u1 > > 2019-10-11 14:59:23 status installed openssl:amd64 1.1.0l-1~deb9u1 > > 2019-10-11 14:59:24 status installed man-db:amd64 2.7.6.1-2 > > 2019-10-11 14:59:24 status installed libssl1.0-dev:amd64 > 1.0.2t-1~deb9u1 > > 2019-10-11 14:59:24 status installed libc-bin:amd64 2.24-11+deb9u4 > > 2019-10-17 16:25:47 status installed sudo:amd64 1.8.19p1-2.1+deb9u1 > > 2019-10-17 16:25:47 status installed apache2-utils:amd64 > 2.4.25-3+deb9u9 > > 2019-10-17 16:25:47 status installed apache2-bin:amd64 2.4.25-3+deb9u9 > > 2019-10-17 16:25:47 status installed apache2-data:all 2.4.25-3+deb9u9 > > 2019-10-17 16:25:47 status installed systemd:amd64 232-25+deb9u12 > > 2019-10-17 16:25:47 status installed man-db:amd64 2.7.6.1-2 > > 2019-10-17 16:25:48 status installed apache2:amd64 2.4.25-3+deb9u9 > > > On Fri, Oct 18, 2019 at 11:12 AM Munroe Sollog <[email protected]> > wrote: > >> Interestingly enough, we started suffering the same problem at the >> same >> time. >> >> - 1 node with 44 cores, 256GB of RAM >> - Zeek 2.5.5 >> - node.cfg: >> [worker-1] >> >> type=worker >> >> host=localhost >> >> interface=af_packet::ens4f0 >> >> lb_method=custom >> >> lb_procs=25 >> >> pin_cpus=0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24 >> >> >> - broctl.cfg: >> >> MemLimit = 100000000 #100GB >> >> setcap.enabled=1 >> >> >> >> On Fri, Oct 18, 2019 at 10:48 AM Mark Gardner <[email protected]> wrote: >> >>> We must have crossed some threshold yesterday. Suddenly we are >>> suffering >>> an epidemic of workers dying with "out of memory in new" even though >>> we >>> made no changes. Previously, we would have a few die each day. Now >>> we have >>> had 250 alerts of workers dying and being restarted from 00:00 to >>> 10:00. I >>> have no idea where to start debugging the problem. Any suggestions? >>> >>> What causes a worker to die by running out of memory? The sensors >>> have >>> lots of memory (see below) so I would not expect to have any out of >>> memory >>> deaths. (To monitor the problem, I am in the process of setting up >>> collectd >>> and graphana.) >>> >>> Some details: >>> - 5 sensors, each with 16-core, AMD Epyc 7351P, 128 GB RAM, Intel >>> X520-T2 >>> - Zeek 2.6.1 >>> - node.cfg: lb_procs=15, pin_cpus=1-15, >>> af_packet_buffer_size=1*1024*1024*1024 >>> - broctl.cfg: setcap enabled >>> - Not shunting any traffic >>> >>> Mark >>> -- >>> Mark Gardner >>> -- >>> _______________________________________________ >>> Zeek mailing list >>> [email protected] >>> http://mailman.ICSI.Berkeley.EDU/mailman/listinfo/zeek >> >> >> >> -- >> Munroe Sollog >> Senior Network Engineer >> [email protected] >> > > > -- > Munroe Sollog > Senior Network Engineer > [email protected] > _______________________________________________ > Zeek mailing list > [email protected] > http://mailman.ICSI.Berkeley.EDU/mailman/listinfo/zeek _______________________________________________ Zeek mailing list [email protected] http://mailman.ICSI.Berkeley.EDU/mailman/listinfo/zeek