[Gluster-infra] regressions running off the rails

[Kaleb KEITHLEY]

On 06/05/2014 10:13 PM, Justin Clift wrote:
> On 05/06/2014, at 5:00 PM, Kaleb Keithley wrote:
> <snip>
>>> I've stopped all tests for the moment. I'm considering rebooting
>>> build.gluster.org, but my track record for it coming back from a reboot is
>>> dismal.
>>>
>>> Avati sent me details for getting on the hypervisor console to kick it if it
>>> doesn't come back from a reboot. I'll see if I can find that but....  it is
>>> getting rather late in the day here for me....
>>>
>>> So if it doesn't come back up, at least you'll know why. ;-)
>>
>> Oh, and I know Justin wants to update _everything_, including OpenSSL to deal with the new vuln. Rather beyond the scope IMO.
>>
>> How do we feel about an general purpose software update?
>
>
> I'm ok if we update just the openssl rpms for now.  As long as we replace this
> box with a new box that isn't so out of date in the near future (weeks).
>

build.gluster.org doesn't use openssl for anything, does it?  We're not 
running https on it AFAIK.  And openssh doesn't use openssl, right?

The machine seems brittle enough as it is. I'd rather focus efforts on 
replacing it with a newer installation with all the latest bits, and 
then regularly updating it.

--

Kaleb