@jkim_wikimedia If you plug it into your laptop and touch the button it will spit out some text. The first 12 characters are your key's ID and the rest is a one-time-use password.

@jkim_wikimedia these are the basic instructions for making an ssh key: https://wikitech.wikimedia.org/wiki/Production_shell_access#Generating_your_SSH_key

Hi @jkim_wikimedia - sorry for the confusion, I'll be making this account for you. Do you have a yubikey?

@Jgreen did this

@BBlack I have been exploring options and it sounds like the DNS TXT record challege would allow us to issue certs without disturbing the hosts. Do you have any caveats about this approach?

He got logged in.

@DStrine what cert problems are you having? First I have heard of that password file problem.

Online and usable at your leisure

Thanks @Papaul and @ayounsi - server is up and accessible.

Looks good, thanks @ayounsi

Typo in config file, updated version: 1538079234

@RLewis you should have received the cert and password, let me know if you can get logged in

@Eileenmcnaughton looks like we already do: https://grafana.wikimedia.org/dashboard/db/fundraising-database?panelId=29&fullscreen&edit&orgId=1&from=1532951492182&to=1537295940406&tab=general

@CCogdill_WMF you have a current cert good until 2019-04-02, the files (ccogdill.pw and ccogdill.p12) are in your home dir on frdev1001. Feel free to reach out on IRC if you'd like a hand retrieving/installing.

@jcrespo thank you for the info! I added the column to frdev1001, and then the error fell to another: log_civicrm_value_partner_19

This is gone, as well as Saurabh and krb5 pass

Apologies, I lost track of this task because I forgot to tag it fundraising-tech-ops

I also see:

Bumping priority because it seems likely payments on Debian 9 is going to fail PCI scans the same way the listener did in T202766

Looks like this was fixed around 7/1

The project is done, shall we remove this access?

Not getting traction at this time

Server is running and creds are in place.

I have asked @jrbs who is chanop there

@Jgreen very good point, maybe something like frack@ ?

I notice that civi1001 is going to double report, as the sender is also fr-tech@

I spoke with user 'grumble' who is freenode staff and they explained how the ban list works, inheriting from #wikimedia-bans. I was able to set +e (ban exempt) in -fundraising and the user rejoined. A cleaner solution would probably be unbanning them from #wikimedia-bans. I don't have op in there.

I think all the failmailing hosts are now reporting: https://grafana.wikimedia.org/dashboard/db/fundraising-overview?orgId=1&panelId=23&fullscreen&refresh=1m&from=now-1h&to=now

I don't even remember what I was talking about, might have been related to https://bugs.python.org/issue12168

@Volans my findings so far:

@Ejegg nope, I tried the IP from my home internet which also timed out - maybe they have added incoming restrictions?

@dbarratt It does look like the biggest images have now been compressed reducing the page load from 10mb to just over 3mb. Thanks!

Seems like there's a lot of room to optimize the images. Even in the US outside big cities one is likely to have too slow/expensive a connection for 10mb page loads.

Yeah I can duplicate, it is running past the (newly doubled) 60s max execution time. My guess is something changed in the code that made this slower because we were getting by with 30s until yesterday.

My current approach is exporting grafana/prometheus data as json and using the SimpleJSON back-end. This will also require:

I adjusted the retention to 6mo for the codfw hosts and it recovered the disk space as expected. Great!

He is set up with kerberos auth for the time being, will be removed when he gets the yubikey

Actually it needs this whole range:

@saurabhbatra96 needs a Yubikey, if shipping is going to cause an inconvenient delay we may be able to work out other authentication in the mean time.

@ayounsi I reverted this since it wasn't the problem after all, please deploy 1531426993 at your leisure

@ayounsi the last one I posted was incomplete, I found the problem and 1531326142 should fix it

With some work the parts of frack puppet necessary for a test bed could be released.

@ayounsi ah yes thanks, I forgot to update the documentation for that, but just did

@awight yes please, also a new fundraising ssh key :)

@schoenbaechler ok you are all set up. If you create the ssh config per https://wikitech.wikimedia.org/wiki/Fundraising/tech/ssh_config you should then be able to log in by typing

works!

@schoenbaechler great! I have created this account, but I forgot to mention you also need a yubikey, do you have one already? If so just paste the output of a press on your user page with the ssh key. Otherwise OIT can get you one.

Site looks good, and slander is working again after kicking syslog a few places

@schoenbaechler in order to get you a login I will need you to make an ssh key, there is info here: https://wikitech.wikimedia.org/wiki/Help:SSH#Creating_a_new_public_and_private_SSH_key

@LeanneS I messaged you on IRC on the 4th, you have a new certificate waiting on frdev1001

@Ejegg @AndyRussG these are now rotated out to /srv/banner_logs, doesn't appear that anything is using the topic yet

After some staring, the existing script is fine