Development for LHC@home

Starting httpd: httpd: Could not reliably determine the server domain name, using 127.0.0.1 for ServerName

Starting vmcontext_epilog ...
bootlogd: no process killed

Is it trying to use 127.0.0.1 as the place to get the x509 credential from ?

No idea, but from what I see here, it's normal and OK.

Well it seems like I have got it working but I have gone back in time again to sitting on a Pacific island beach...

Thu Sep 10 07:44:34 2015: Starting rpcbind: ^[[60G[^[[0;32m OK ^[[0;39m]
Wed Sep 9 23:44:52 2015: Starting CernVM: ^[[60G[^[[0;32m OK ^[[0;39m]

Working on the idea that you may be having problems connecting to whatever it is that supplies the credential, a couple of thoughts on things that can affect the ability to connect to some networks. They come with more caveats than you could shake a stick at, but here goes:-

1. A problem with a DNS cache. There will be one in each host. Although unlikely, it's possible that they're all affected, especially if the cause is malicious. I don't know what OS you're using, but if it's Windows you can flush the cache using ipconfig/flushdns (different Windows might have different commands but there will be one somewhere). There will also probably be a cache in the router. The DNS settings in the network configuration of each host will show if you're using this. Presumably a factory reset as suggested here will flush this. You could try different DNS servers, too, especially in the router setup, if used.

2. Router MTU settings. A discussion on one of the forums of my ISP mentioned this as causing problems connecting to some networks. This is the help text from a setup screen of my router:-

"The normal MTU (Maximum Transmit Unit) value for most Ethernet networks is 1500 Bytes, or 1492 Bytes for PPPoE connections. For some ISPs you may need to reduce the MTU. But this is rarely required, and should not be done unless you are sure it is necessary for your ISP connection."

The default is 1500, if you're using PPPoE and are sending 1500 byte packets, they will be fragmented and some (distant end) network setups apparently don't like this. For example, my firewall has an option to drop them. I've tried changing mine from 1500 to 1492, both work for me (it's back to 1500 now) but it might be worth asking your ISP what value you should use, or just try changing it. Maybe fibre makes a difference. If you set it too low you'll reduce the throughput.

My ISP has been having DNS problems, big one this morning.

What I did this morning to make it work (on a Windows box) was add some non-ISP Tier 2 DNS servers. I then went and tried on another Windows box (without adding extra DNS servers) and that worked too, I decided the router cache might have been why it worked for that one too.

I have Windows and Ubuntu, mostly Ubuntu at the moment but can swap them over when needed.

My router died this afternoon, barely a week old ! Have had to go and buy a new one as the ISP replacement probably won't arrive till Monday.

The 2 jobs I was running today have picked up new work after I plugged and configured the new router this evening.

I'm thinking DNS is/was the problem but others have also reported the same error but for them they quickly get more work when mine would fail completely once I saw the failure to get an x509 credential.

I have no idea why suddenly in the middle of setting everything up at the beginning my time clock jumps back those 8 hours (7 hours back plus 1 hour BST). It never used to, then it did, then it didn't, now it does !!!