Evergreen ILS Website

IRC log for #evergreen, 2021-10-11

| Channels | #evergreen index | Today | | Search | Google Search | Plain-Text | summary | Join Webchat

All times shown according to the server's local time.

Time Nick Message
03:30 JBoyer joined #evergreen
03:30 jweston joined #evergreen
03:30 miker joined #evergreen
03:30 akilsdonk joined #evergreen
06:00 pinesol News from qatests: Testing Success <http://testing.evergreen-ils.org/~live>
09:18 collum joined #evergreen
09:27 mantis joined #evergreen
10:35 Dyrcona joined #evergreen
10:36 Dyrcona Anyone else had fun with systemd-resolved after updating packages on Ubuntu 18 lately, or is it just me?
10:36 Dyrcona The problem is that I also updated and rebooted my load balancers, but I don't think it was anything in their config that was lost.
10:43 Bmagic not yet, but maybe your my canary
10:48 Dyrcona It all worked before doing apt upgrade and a reboot.
10:49 Dyrcona What makes me suspect iptables on the load balancer is it works on machines with external IPs as well as non-routable IPs.
10:49 Dyrcona I have tested on all of my vms, yet.
10:50 Dyrcona That should be I have NOT tested on all of the vms.
10:50 Dyrcona It's always something...
10:52 Dyrcona OK. It is not iptables. DNS works on a physical server that was upgraded previously from Ubuntu 16 to Ubuntu 18, had updates applied today, and still has /etc/network/interfaces. So, maybe its with netplan and systemd?
10:52 Dyrcona The DNS nameservers are the same.
10:53 Dyrcona Grr. brick 4 is out of rotation....
10:54 Dyrcona Never upgrade. It isn't worth it.
10:54 Dyrcona Grr. technology--
10:55 Dyrcona Just login to brick 4, and it's magically working again.
10:57 Dyrcona connection timed out; no servers could be reached # Makes it look like a networking issue....
10:59 Dyrcona So, it works with actual hardware and VMs that have both a 192.168 and external IP, but not whose with only a 192.168 address. Must be networking on the hosts. They got updated and rebooted today.
11:02 Dyrcona The hosts' DNS lookups fail.
11:03 Dyrcona OK, right. The hosts fire up dnsmasq anyway, so something must have changed there.
11:04 stephengwills joined #evergreen
11:04 stephengwills left #evergreen
11:12 Dyrcona Why is it never the problem from StackOverflow? No.... My problems always have to be "special."
11:13 Dyrcona An, it looks like dnsmasq is pretty much only doing dhcp for libvirt on the host I'm trying to fix.....
11:16 Dyrcona And, sure, journalctl says nothing wrong with systemd-resolved.....
11:25 Dyrcona FYI: If you're DNS is busted, book jackets don't work.
11:26 Dyrcona s/you're/your/
12:37 collum joined #evergreen
12:43 collum joined #evergreen
13:15 mantis This was something I asked once already in IRC.  I tried to set up an autorenewal notice in a test environment to display the item barcode but received failures.  The latest I had were  [%- SET idx = loop.count - 1; SET udata =  user_data.$idx -%]
13:15 mantis [%- SET cid = circ.target_copy.id || udata.copy -%]
13:15 mantis [%- SET copy_details = helpers.get_copy_bib_basics(cid) -%]
13:15 mantis Barcode: [% circ.target_copy.barcode %]
13:17 Dyrcona mantis: You added circ.target_copy to the environment: action_trigger.environment table?
13:19 mantis ah I didn't I'll try that now
14:03 Dyrcona iptables--
14:03 Dyrcona pf++
14:18 Dyrcona So, my DNS problem ended up being iptables on the load balancer/router. I had to switch from a SNAT rule to a MASQUERADE. I'm not sure why the SNAT stopped working for DNS, but still seemed to work for connections via IP address. Oh well, it's all working, now.
14:25 Dyrcona On the plus side, we're running Pg 10 in production. The pg_upgrade with a copy, not --link, took about 1 hour, then another hour or so to restart replication with pg_basebackup.
18:00 pinesol News from qatests: Testing Success <http://testing.evergreen-ils.org/~live>
21:35 degraafk joined #evergreen

| Channels | #evergreen index | Today | | Search | Google Search | Plain-Text | summary | Join Webchat