-
jesse_
bahamat, the start will succeed, after some indeterminate time
-
jesse_
so there is some timeout that will handle it, eventually
-
jesse_
cutting from the status:
-
jesse_
nginx:
-
jesse_
Active: active (running) since Wed 2022-10-12 01:50:41 UTC; 4s ago
-
jesse_
Process: 67771 ExecStartPre=/usr/sbin/nginx -t -q -g daemon on; master_process on; (code=exited, status=0/SUCCESS)
-
jesse_
Process: 67772 ExecStart=/usr/sbin/nginx -g daemon on; master_process on; (code=exited, status=0/SUCCESS)
-
jesse_
postgresql:
-
jesse_
Active: active (exited) since Wed 2022-10-12 00:50:44 UTC; 57min ago
-
jesse_
Process: 63727 ExecStart=/bin/true (code=exited, status=0/SUCCESS)
-
jesse_
whops, missed main pid line:
-
jesse_
nginx: Main PID: 67773 (nginx)
-
jesse_
postgresql: Main PID: 63727 (code=exited, status=0/SUCCESS)
-
jesse_
the difference must be nginx being 'stay running' and postgresql being 'fork a background process'
-
jesse_
is there a defined process to handle postgres databases when doing native zone upgrade (with delegated dataset)?
-
gonzosysadm[m]
since when is piadm (command) a thing? pretty useful stuff
-
gonzosysadm[m]
by the way, any freebsd13 images floating around? failing that, is the freebsd12 one good?
-
gonzosysadm[m]
is there any configuration i can add to silently drop connections to closed ports on the GZ? like net.inet.tcp.blackhole=2 on freebsd
-
nikolam
wow, i get ** server can't find pkgsrc.smartos.org: SERVFAIL for pkgsrc.smartos.org on my local internet provider.. (major one)
-
nikolam
And sometimes it does resolve ti, but in like one in 10 tries.. May be loca issue with provider NS or..
-
neuroserve
8.8.8.8 answers promptly
-
nikolam
also does 1.1.1.1 .. but local one..
-
travisp
gonzosysadm[m]: I haven't had any issues with the freebsd 12 image though I also haven't used it very much. I did upgrading it to 12.3 without issue.
-
travisp
gonzosysadm[m]: The freebsd 12 image appears to work well, I haven't used it heavily but I was able to upgrade it to 12.3.
-
travisp
gonzosysadm[m]: sorry for the duplicate send, it looked like Thunderbird ate the first message.
-
barfield
bahamat: should dehydrated be installed on all headnodes in order to work on ha instances such as cmon?
-
bahamat
It only needs to be on the primary headnode. It knows how to distribute the cert to each cmon instance.
-
barfield
Weird, I created a couple of HA cmon instances but they still have self-signed .triton certs.
-
barfield
depth=0 CN = *.triton
-
barfield
verify error:num=18:self signed certificate
-
bahamat
-
barfield
Ha
-
barfield
Thanks!