Hey Lee, I figured out the problem..... It was actually the VPN that terminates the connection after one hour. The entire backup simply took too long, so the connection was considered idle. We've switched to parallel dumps now and Capistrano no longer looses the progress and everything works as expected.
Kind regards, Arne Op donderdag 20 oktober 2016 18:40:10 UTC+2 schreef Lee Hambley: > > I suspect you're right that the task is hanging because of a lack of > output, it's a common issue. > > Is there some settings we could look into? >> Like a TTL, Timeout or have the task report back the progress/output from >> the pg_dump or something? > > > first thing that comes to mind is to run the output of `pg_dump` through > `pv` [1] and set it to print some periodic status update, pv is a pretty > worthwhile little tool anyway, and it might solve this problem really > neatly. > > [1]: Pipe Viewer (http://www.ivarch.com/programs/pv.shtml) > > Lee Hambley > http://lee.hambley.name/ > +49 (0) 170 298 5667 > > On 19 October 2016 at 12:38, Arne De Herdt <[email protected] > <javascript:>> wrote: > >> Versions: >> >> - Ruby 2.3 >> - Capistrano 3.6.1 >> - Rake / Rails / etc 11.3.0 / 4.2.7.1 >> >> Platform: >> >> - Working on OSX >> - Deploying to Ubuntu/CentOS >> >> >> Hello, >> >> >> We are currently running into a weird behaviour with Capistrano that >> started appearing when our database backup exceeded the 4GB in size. >> >> I've attached the rake task for generating a backup to show you how it >> works, it's basically creating a pg_dump. >> >> >> The problem we encounter, is that before doing the deploy, we ask whether >> a database backup is needed. >> >> If the answer is yes, then we invoke the task that is in the backup.rake. >> >> Now when the backup is larger then 4GB, Capistrano seems to lose the >> connection with the server. >> >> We do not get any feedback from Capistrano whether the task succeeded, >> failed or is still running. >> >> So after roughly one hour, we kill the task locally by pressing ctrl +C, >> and run the deploy again without backup. >> >> >> The result is that we then succeed at deploying, and the backup is >> actually being made. >> >> So we know that the backup task itself actually works, it's just that >> Capistrano doesn't send us feedback. >> >> >> Is there some settings we could look into? >> >> Like a TTL, Timeout or have the task report back the progress/output from >> the pg_dump or something? >> >> >> Kind regards, >> >> Arne De Herdt >> >> -- >> You received this message because you are subscribed to the Google Groups >> "Capistrano" group. >> To unsubscribe from this group and stop receiving emails from it, send an >> email to [email protected] <javascript:>. >> To view this discussion on the web, visit >> https://groups.google.com/d/msgid/capistrano/5bc45616-432f-4ab1-9907-39a076d95f46%40googlegroups.com >> >> <https://groups.google.com/d/msgid/capistrano/5bc45616-432f-4ab1-9907-39a076d95f46%40googlegroups.com?utm_medium=email&utm_source=footer> >> . >> For more options, visit https://groups.google.com/d/optout. >> > > -- You received this message because you are subscribed to the Google Groups "Capistrano" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. To view this discussion on the web, visit https://groups.google.com/d/msgid/capistrano/0cf320e3-3d7c-4d42-862f-6887bb01e078%40googlegroups.com. For more options, visit https://groups.google.com/d/optout.
