[mythtv-users] Random Hard Frontend Lockups

[Hika van den Hoven]

Hoi R.,

Friday, March 27, 2015, 3:08:40 AM, you wrote:

> On 26/03/15 07:21 PM, Hika van den Hoven wrote:
>> Hoi Hika,
>>>> I *am* beginning to think it may be a memory fault, in the space where X
>>>> or the frontend will always reside. (ie: 'above' the backend/mysql
>>>> processes which are started before X is launched.)
>>
>>>> Geoff
>>
>>
>> I'm thinking about another possibility. Am I right that even when
>> residing on the same machine backend and frontend communicate through
>> a net socket? Could it have to do with an intermittent failure on the
>> nic? Easy to check if you have a spare by replacing it.
> The system runs well for hours, and locks up, in the middle of the 
> night, when the only thing going on would be comm-flagging.
> I was under the impression that when a system is in fact using 127.0.0.1
> to talk to 127.0.0.1, then the socket is aliased into file space and IP
> is bypassed. ( I might be thinking about how mysql works and not other
> more usual IP stuff..)

> The NIC is a fairly new standalone card. It could be bad. But when the
> internal NIC went, it was instant and total

> G.

I don't know either, but I was thinking so because to mysql 127.0.0.1
is not the same as localhost. If I remember right localhost bypasses
net socket and 127.0.0.1 not. And if you have your myth system set to
the nic address ...
I'm just theorizing, because some faulty memory doesn't sound right
with the backend continuing and at the same time nothing in the log?!
If the kernel is sane enough to let the backend continue, it should
also be sane enough to notice something and put it to the log.
Especially if it's memory corruption.

Tot mails,
  Hika                            mailto:hikavdh at gmail.com

"Zonder hoop kun je niet leven
Zonder leven is er geen hoop
Het eeuwige dilemma
Zeker als je hoop moet vernietigen om te kunnen overleven!"

De lerende Mens