#7306 IRC channel logging for #fedora-coreos
Closed: Will Not/Can Not fix by dustymabe. Opened by dustymabe.

  • Describe what you need us to do:

Help us come up with a plan for logging #fedora-coreos IRC channel. We have discussed this with legal@ and would like to follow their recommendations of:

- limit logging to 2 weeks of time
- notify users clearly of "round the clock" logging (topic and welcome message)
- have a bot that reminds users frequently of channel logging 
- work with fedora infra to set the infrastructure up

The botbot.me software is open source and we like the interface. Seems like maybe a good candidate for an openshift pod/service. WDYT?

related: Fedora CoreOS issue on this topic

  • When do you need this? (YYYY/MM/DD)

preferrably by 2018/11/05 when botbot.me shuts down

  • When is this no longer needed or useful? (YYYY/MM/DD)

NA

  • If we cannot complete your request, what is the impact?

We find some other way to log.


So, I noticed couple of things here, botbot is going to shutdown on 5 November of this year. I suppose we could clone the git, and make a fork of it, but then we need to maintain it (we here is definately the queens we.)
Also, we've got meetbot in some channles, notabily the fedora-meeting channels, which takes the meetings, once started, logs them, and feeds them into the log site. I don't think the feed is currently a live stream out to the log site, and it only gets parsed when the meeting ends. I also know that we have moved from supybot to limira, or something similar. I've not looked at all* into this, but possibly we could figure out a way to make it so meetbot can pipe out to a file in realtime. We'd then have to find a way to make the website update in real time, which is something I have never looked at here. In short, have something that might work already, with a bit of tweaking, so do we really want to look into, at this point, taking over and forking something that is about to go the way of the dodo, thanks to GDPR?

Lastly, i'm a little fuzzy on how the around the clock logging has a limit of two weeks, or are they saying only two weeks at a time? As far as reminders go, we could in therory make a bot do that, or kick a meetbot session every hour or so, so it is popped into the channel stream that we are being logged.

Metadata Update from @bowlofeggs:
- Issue priority set to: Waiting on Assignee (was: Needs Review)

@bowlofeggs - was this erroneously put into Waiting on Assignee state ? There is no assignee.

@dustymabe No, we use that state to mean that the ticket has been triaged and is waiting on someone to do it (i.e., it is waiting to have an assignee).

ahh. ok. I see.

at this point it is on ticket creators to do a POC and provide working deployment to the infra team to collaborate on setting this up.

Hello. Reading through the posts here, am I correct in thinking you all are actually using Limnoria currently? If so, I believe I could try to write a plugin for the proposed capability. Limnoria's "ChannelLogger" plugin does contain log rotation capability already, but it doesn't contain any way to roll the logs. I think that's what I could try to flesh out. Maybe rotate logs once per day, and roll them every 14 days? Couldn't be too terribly hard.

Yes, zodbot is indeed Limnoria. Most likely version 20171025 from EPEL7.

I don't think anyone would turn down an implementation if you were so kind as to provide one.

great, thank you. I'll see what I can come up with.

I seem to be unable to assign myself to this issue though. is this something you or someone else would be able to do?

Can someone please remove me from this ticket completey=3F I've decided t=
o leave the project with no intention to return.

David Shier (odin2016)
Gesendet von Mailspring (https://link.getmailspring.com/link/1550029315.l=
ocal-6f6f441a-8333-v1.5.6-4cb1851b=40getmailspring.com/0=3Fredirect=3Dhtt=
ps%3A%2=46%2=46getmailspring.com%2=46&recipient=3DcmVwbHkrNWJhN2YxM2U0MTA=
1YjllMD=46mNzVmMjQ2ZDY2YTM2MzA2M2YwNmY5YjNiY2E5ZWJlOTI5YTcyNTA0Nz=46jNmE3=
NDBiZDc2NWExYjI5Zjk4NmJkM2ViOTAxMGMyNj=46hOWE0MGQ2NDE5YWVkNjYwMzQ5ZGY3NzZ=
mOWQ0NjcyOThiNDlAcG=46ndXJlLmlv), der besten kostenlosen E-Mail-App f=C3=BC=
r die Arbeit
On =46eb. 12 2019, at 10:21 pm, Christopher Webber w=
rote:

cwebber added a new comment to an issue you are following:
=60=60
great, thank you. I'll see what I can come up with.

I seem to be unable to assign myself to this issue though. is this some=
thing you or someone else would be able to do=3F
=60=60

To reply, visit the link below or just reply to this email
https://pagure.io/fedora-infrastructure/issue/7306

Can someone please remove me from this ticket completely I've decided to leave the project with no intention to return.

You can unsubscribe yourself from the ticket by clicking on the "unsubscribe" button on the right hand side column

We've found a third party service called echelog that agreed to log our channel for us. I'll confirm once they set it up (should happen tomorrow) and then this ticket can be closed.

The channel is now being logged by echelog. This ticket can be closed.

@dustymabe Is logging by echelog enough, or is there still a need to have logging set up by Fedora infrastructure?

Metadata Update from @mizdebsk:
- Issue priority set to: Waiting on Reporter (was: Waiting on Assignee)

@dustymabe Is logging by echelog enough, or is there still a need to have logging set up by Fedora infrastructure?

logging by echelog is enough. we simply need to be able to reference conversations that had when people weren't in the channel or for users themselves to be able to go back to conversations they've had in the past. While we don't have a requirement for now for this (since echelog is taking care of it) it might happen again where we lose our provider and circle back on this.

closing out for now.

Metadata Update from @dustymabe:
- Issue close_status updated to: Will Not/Can Not fix
- Issue status updated to: Closed (was: Open)

Metadata