If anyone has noticed any disconnects over the last hour or so, it seems there's some mineor internet problems around the pacific.
I've not noticed the pool hash rate change any more than usual, but a linode monitoring node keeps complaining about node and pool connections.
Very few miners should be affected, however, and they should just be reconnecting pretty much as soon as it happens.
Would this have anything to do with all my S9's going offline? I got out of bed an hour ago and noticed that all my S9's had little xxxxxxx all the way across on all hash boards. I rebooted them all and they're back up, could it be pool connection related somehow? None of my S7-LN were affected.
I don't think xxxxx boards have anything to do with the pool. If there was a major disconnect to the pool, then ALL of your miners would be affected, not only s9's. Besides, your s9's would be mining at your backup pool. But I've read here and in some other threads that s9's do have some internal issues. I have only s7's so can't really say what could be the cause in your case.
Yeah thats a point, it was just so weird because all my miners are hooked into the same power, same internet, same everything. So, theoretically if power slipped, they should have all taken a dump, pool crapped out, all should have dropped, etc. I'm gonna keep an eye on it, but it is quite odd... Trying to stop myself from putting on a tin foil hat haha
I've heard of this behavior from others, but no clear indicators for the reason (other than what BM probably already knows). I still think it can be a power issue, aside from heat. I don't use BiteMe PSUs. When I began using S3s a while back, I got "stuck" on the HP 1KW PSU bricks for servers, using gridseed adapters with heavy wire. It takes two of them to run an S7, or an A6, or an S9. I ran two boards off of one brick, and the other board and controller off of the other for an S7. I didn't buy an S9, but figure it's probably close. The HP bricks run very efficiently, and using that kind of configuration results in no PSU having to put out more than around 75%. Anyway, it could be that your S9s are just that much more sensitive to 12v level, and either a shaky PSU, or line power that fluctuates and consequently (except in higher end PSUs that compensate for that adequately) may vary the output power, and that issue may not show up on other gear. We have a shaky power grid here on the island...that's an issue here, and we all use line conditioners or high-end PSUs because of it. It can also happen because of old wiring and so-so grounds. I suppose that's not very helpful, but seemed I should share it.
EDIT: What brought this to mind is that when I was running all six S3s, they were overclocked to varying degrees. An S3 has four PCIe plugs; it only needs two (one per board) to run normally. On some, but not all, of them if I went above default clock, I'd start getting xxx's even though the temps were still OK. When I'd plug in the other two PCIe inputs, the x's would disappear.
