I don't know if other miners will work without any modifications, but poclbm have some problems with DNS resolving, so it don't work yet. But I'll try to fix it as soon as I find some time when pool will be stable enough :-)

Ok, now is on all backends kernel compiled directly for virtualization. Give it a try!

Hm, it looks like serious issue with linux kernel and vmware. Pool is down, I need to find a solution. I'll probably try older kernels, which are reported as working.

I think that my miners have some driver issues, but there might be more reasons.  Yesterday one my poclbm stopped working and told me somethign about hardware issue, but I know that trigger was shutting down of the pool. So miners can be sometimes really weid :-)

Sargasm: Didn't they lock up during pool outage for some time? It happen sometimes on my miners. However i'm pretty sure they were affected by recent pool instability, which should be OK now.

Btw calculated shares are not (directly) related to score, so this means those workers really didn't sent accceptable shares to pool.

I had some other stuff until now, but I'm going to payments soon, should be up to two hours.

I forgot to say that Tor mining (although it's still highly experimental feature) is fully working on new server, including long polling support and ntime rolling. I didn't finished patch for poclbm yet because recent server issues, but if anybody want to play with his miner and Tor support, it's done on pool side.

I'm also surprised by low latency on Tor network. I remember than few years ago it was nearly impossible to browse onion sites, but now it's pretty comfortable. Of course building circuit take some time, but then I'm usually having responses under one second, which is excellent.

New Tor address for pool is: poolziyzjb25aa4e.onion pool57wkuu5yuhzb.onion

Problems in long polilng connection. You get few of this when I restart long polling daemon, but your miner should then reconnect. I don't see any issue with longpolling on my side now.

I'm an idiot, because I has been trying another mysql configuration on separate virtual machine - and forgot to setup same kernel parameters there as on production database. So I accidentally shot down machine on the same vmware bug again. So outage was just a server restart, because I was on console when it happen. I'm sorry for downtime.

bulanula: Payouts are not enabled yet, they'll be back in few hours. I need to check last rounds if their rewards were not affected by outages and recalculate possible issues. Thanks for patience.

I had nice time with OVH technical support, it was really pleasure to work with them. The best part is that it was NOT a hardware issue, but a bug in VMWare which *looked* like hardware issue, which has been fixed by tuning kernel options. This also explain why RAID "didn't work". Really painful 17 hours of waiting to their response! However it's fixed on 90% now.


Well, to be honest, they did setup RAID on the server, but it isn't visible on software layer (and even in their disk recovery tool in rescue mode), so I decided that RAID is missing by mistake.

No coins were lost, I'm doing regular backups outside datacenter and I did also manual backup before shutting down the server. However payouts are now stopped, I need to recalculate few rounds manually before I'll start them again, sorry.

I worked with EC2 in my previous job, of course I considered it. Unfortunately they handle DDoS in similar way like Linode.

Do anybody plan to sell Casascius coins on conference? It would be nice to buy few of them without paying overseas post...

^ this

Btw what's the probability that new server will fail second day in production, when I tested it almost a week? That's insane "luck".

Firstly, I had some good recommendation for them. Then I ask them directly how they handle DDoS and their response was pretty satisfying, they promise that DDoS attack won't be a reason for cancelling our agreement. Last thing is also that price for current server (which is mostly high-end; multi CPU, plenty of RAM, 10Gbit uplink, SSD RAID) was very good. They have officially 1 hour GTI and 2 hours GTR (time to repair issues), however they're totally failing now (which is hard to know in advance, unfortunately).

Btw I has been in touch with Rackspace, I even had servers here and hardware firewalls for fighting DDoS, for really insane money. But then I realized that even their "DDoS solution" is shitty for such large scale attacks which we have in bitcoin world, so I hanged it. And it was good choice, because their monthly price was *many times* higher than pool income at this days.

OK, it failed again. Sorry, there's no other chance than wait for crappy OVH support to replace the disk. Damn! :'-(

I'm recommending you to go solo or join another pool, because pool will be down until tomorrow.

Edit: hooray, they started responding! Hope it gets better sooner than tomorrow.

Well, diagnosis is behind us, they're preparing new HDD. Unfortunately it's France, so it will be ready probably tomorrow.

Pool is now back up, I did some prevention like marking some parts of disk as corrupted, so I hope it will be enough until tomorrow.

I'm doing my best, unfortunately OVH support did some mistakes like they forget to setup RAID during server installation (heh). Now I know that I need to re-check everything they're doing.

I'm assuming that you're using correct IP (which has been changed in last two days); then yes, I'm sometimes experiencing crash on my miners, too. I think it is caused some bug directly by miner or GPU drivers, because sometimes I need to restart machine completely. It's pretty annoying, because such outages like today can turn off many workers.

Ok, I'm turning down the pool because needed hardware diagnosis. This is really sad and I'm insanely upset. That's NEW SERVER.