Got fixed somehow. Might be related to that instance getting corrupted at one point.

reopening, wrong bug.

So that is indeed the new instance but it does not let us in :(

Running the ssh proxy command with verbose gives:

debug1: channel_connect_stdio_fwd deployment-bastion.pmtpa.wmflabs:22
debug1: channel 0: new [stdio-forward]
debug1: getpeername failed: Bad file descriptor

Indeed, there is no host entry for deployment-bastion :

hashar@bastion1:~$ host deployment-dbdump
deployment-dbdump.pmtpa.wmflabs has address 10.4.0.56
hashar@bastion1:~$ host deployment-bastion
Host deployment-bastion not found: 3(NXDOMAIN)
hashar@bastion1:~$

Changed summary to: "labs DNS lacks an entry for deployment-bastion.pmtpa.wmflabs"

Need ops to create the DNS entry.

Not terribly urgent so lowering priority. We are still using deployment-dbdump so a possible way to fix the issue is to just recreate the instance. I let it around so ops could eventually find out what failed when inserting the DNS entry.

I have killed the instance and recreating it.

Still the same issue, so we need ops to update the DNS entry.

This has been fixed over the month of Augsut 2012.