On August 3, 2020 12:00:41 AM GMT+03:00, Todd Benivegna <t...@benivegna.com> wrote: >How would I capture with Wireshark when it seemingly happens at >random? Whenever I test, everything appears to work normally. I wish >I could somehow replicate it. Maybe I need to test manually and see if >it’ll happen in front of me; I’ve never actually seen it happen… always >happens when I’m away. > Just let it run filtering by host and port and logging to a file
>-- >Todd Benivegna // t...@benivegna.com >On Aug 2, 2020, 12:12 PM -0400, Manuel Wolfshant ><wo...@nobugconsulting.ro>, wrote: >> On 8/1/20 11:25 PM, Todd Benivegna wrote: >> > I'm hoping that someone can shed some light on this… I have a >Synology NAS (DS416) that has a feature where you can enable a “Network >UPS Server” which is a NUT server. I have been trying to get the >Synology to shut down three Ubuntu 20.04 servers that I have. While it >does work when I test it out manually, sometimes when I am away and the >power goes out briefly, the servers shut down when the power has been >out for like five seconds (or less sometimes). When I come home and >turn on the servers, they boot up but then immediately shut down >again. This happens until I restart the Synology; then they will boot >up and stay up. I‘d like them to stay up a little longer than >that. Ideally, I’d like them to stay up until battery is low, then >shut down, then all come back on when power is restored. >> > >> > This is what I have done so far: I have enabled "Enable Network >UPS server" on the Synology and have installed NUT on each of my >servers running Ubuntu 20.04. I have added the appropriate IPs to the >Permitted DiskStation Devices” list. I have also tried setting it on >the Synology to shut down when battery is low and after a specified >amount of time (20 minutes). Either way, the servers will shut down >after like 5 seconds or less. I have edited upsmon.conf and added my >MONITOR line and setup systemctl so that the nut-client service starts >automatically at boot. I have no made any other changes to the file; >the rest is still set to defaults. >> > >> > So I'm not sure where exactly the problem is; if it's the Synology >or NUT on Ubuntu. Strange thing is, when I manually test by shutting >off the power briefly (or for a few seconds, or a few minutes - I've >tried everything;) every time I do a test, everything works perfect and >they will shut down when they are supposed to. Seems to only happen >when there is a passing storm that knocks the power out for a few >seconds. >> > >> > Also, this is what I found in the syslog on one of the machines: >> > Jul 31 18:33:29 plex upsmon[970]: UPS ups@192.168.1.70 on battery >> > Jul 31 18:33:34 plex upsmon[970]: UPS ups@192.168.1.70 on line >power >> > Jul 31 18:34:04 plex upsmon[970]: Executing automatic power-fail >shutdown >> > Jul 31 18:34:04 plex upsmon[970]: Auto logout and shutdown >proceeding >> > Jul 31 18:34:09 plex systemd[1]: nut-monitor.service: Succeeded. >> > >> > If I’m not mistaken, it is shutting down after power came back on >line….? >> My first suspect is the Synology version of nut. More specifically, I >suspect that nut triggers a shutdown immediately after the switch to >"on battery" state and only cancels it after a restart. >> What I would do is to use wireshark to sniff the communication >between the Synology and one of the Ubuntu machines and log what >exactly ( and when ) is sent by the nut server to the clients. I'd also >couple that with a bit of logging on the nut side as well ( for 60 secs >or so, starting shortly before the switch to on-battery, with a very >short interval between polls ) >> _______________________________________________ >> Nut-upsuser mailing list >> Nut-upsuser@alioth-lists.debian.net >> https://alioth-lists.debian.net/cgi-bin/mailman/listinfo/nut-upsuser _______________________________________________ Nut-upsuser mailing list Nut-upsuser@alioth-lists.debian.net https://alioth-lists.debian.net/cgi-bin/mailman/listinfo/nut-upsuser
