Dub

Members
  • Posts

    9
  • Joined

  • Last visited

Dub's Achievements

Noob

Noob (1/14)

0

Reputation

  1. I hate to necro this, but did anyone else come up with a solution for this? I've also deleted the audio codec and still have this issue. I even setup a job to just do this on the regular and restart the docker and I still get these filling up the entire log: Error while decoding stream #0:1: No space left on device They just fill up the entire log and pegs the CPU, even when nobody is watching. Plex cranks up the CPU usage and maxes out for hours for no reason like this: Again - I've done the audio codec purge on the regular, it still does it after a reset once it sits idle for a spell. I've rebuilt the entire docker, all stock, default transcode config: Space is not an issue on the container. in my case it has 30gb free in the /transcode directory. I thought maybe it was a writing issue, so I even tried transcoding directly into ram and that didn't solve it either, so I went back and just use /tmp. If anyone has any insight, that would be amazing.
  2. For anyone finding this later, Supermicro makes the H12SSL-CT which is also 7003 compatible EPYC, has dual 10G nic's, 2x m.2 NVME, and has a broadcom controller on board.
  3. Somewhat a rehash of my previous post I'd inadvertently put in the wrong forum as I didn't realize there was a dedicated motherboard forum. I've given up on finding the original configuration I'd planned and just resigned myself to getting a working Intel solution in the interim. Ideally I'd love suggestions on a Supermicro Intel board + CPU combo with (hopefully) at least one 10G nic and at least 1 x16 & 5 x8 PCIE slots that would at least be as good single-thread wise as my 3950x (cpumark of ~2738) would be fantastic. If it has m.2 slots that's also a big plus - but if it has 10G nic's I could put in a PCIE m.2 adapter if I had to. Any suggestions are appreciated!
  4. I've been prepping to upgrade my server to support more drives and just as I was about to snag all the bits, I can't find the ROMED8-2T motherboard I'd planned to use in stock anywhere for a reasonable price and it seems to have doubled in price everywhere in the last two weeks. I was able to snag a CSE-846 and am expanding from 6+2 to 16+2 /w ideally space for up to 22+2 /w 2 mirred NVME cache drives. I'm currently using an AM4 board but am maxed out on PCI slots to add the additional 2x LSI 9201-8i's to support the additional drives. The backplane in the CSE-846 is a direct pass-through and not an expander so I need direct connections for each drive which is one of the big reasons I was swapping over to the ROMED8-2T. Primary purpose is storage + Plex (lots of hardware encoding using quadro). Ideally I'd love for the board to have a BMC and support at least 5 PCI slots + 2 NVME slots. Doesn't have to be AMD. On-board 10GB NIC would also be icing on the cake and save me another slot. If there's a supermicro alternative that would drop right in, that'd be even better. If you have any suggestions, I'd love to hear them!
  5. I was able to recently snag an old CSE-846 that has an existing SAS846A backplane, and was also able to nab an LSI 9750-24i4e at the same time. I understand the original SAS846 (non A) wasn't able to support > 2TB drives but I was told the A variant should be fine. I won't have the drives for a little bit - but after trying to find some info on the card and the backplane I'm curious if I'm going to have issues with either of those and if I should be looking at other options before I start migrating to the new hardware.
  6. I'm running out of disk and was considering upgrading to a Fractal 7XL to be able to hopefully double drive space. I originally went with consumer-level hardware just because I have a Quadro P2000 and wanted to keep using it primarily because I have a lot of folks transcoding 4K. I assumed I'd need to bump up to a server case but when I look at server cases most don't support full-height cards or require weird specialized motherboards specific to a case - which I can't seem to ever find all the appropriate pieces (e.g. SM 847E16-R1400UB). Because I went with AMD I don't have any more PCI slots to add another HBA due to having the Quadro, an HBA card, and a 10gbe card. The end configuration needs to support at least 16 spinning drives + 2 NVME drives. More SATA for a little future-proofing would be a bonus. Would anyone have motherboard suggestions that would fit a 7XL that have at least 4 16x PCI slots or a motherboard that preferrably has 10GBe and/or a server motherboard with at least 2 mini-sas connectors (and preferrably the 10GBe)? OR Suggestions for a 24bay/4U style enclosure that would actually support a full-height quadro and appropriate motherboard? I don't have a specific budget in mind, just trying to figure out the best option. Thank you in advance!
  7. Apologies, I didn't realize the entire diagnostics would be necessary. diagnostics-20220106-0327.zip
  8. I find lots of suggestions on how to resolve a parity situation when you have one parity drive, but no specifics when you have more than one. I'm unsure if anything is different about the process. I recently had one of my two parity drives fail and I'm apprehensive about doing much of anything without a little guidance. Here is what SMART is throwing: ATA Error Count: 2 CR = Command Register [HEX] FR = Features Register [HEX] SC = Sector Count Register [HEX] SN = Sector Number Register [HEX] CL = Cylinder Low Register [HEX] CH = Cylinder High Register [HEX] DH = Device/Head Register [HEX] DC = Device Command Register [HEX] ER = Error register [HEX] ST = Status register [HEX] Powered_Up_Time is measured from power on, and printed as DDd+hh:mm:SS.sss where DD=days, hh=hours, mm=minutes, SS=sec, and sss=millisec. It "wraps" after 49.710 days. Error 2 occurred at disk power-on lifetime: 10463 hours (435 days + 23 hours) When the command that caused the error occurred, the device was active or idle. After command completion occurred, registers were: ER ST SC SN CL CH DH -- -- -- -- -- -- -- 40 51 00 b8 96 01 0e Error: WP at LBA = 0x0e0196b8 = 234985144 Commands leading to the command that caused the error were: CR FR SC SN CL CH DH DC Powered_Up_Time Command/Feature_Name -- -- -- -- -- -- -- -- ---------------- -------------------- 61 00 00 ff ff ff 4f 00 19d+01:46:18.556 WRITE FPDMA QUEUED 61 00 00 ff ff ff 4f 00 19d+01:46:18.509 WRITE FPDMA QUEUED 60 00 c8 ff ff ff 4f 00 19d+01:46:18.508 READ FPDMA QUEUED 60 00 00 ff ff ff 4f 00 19d+01:46:18.508 READ FPDMA QUEUED 60 00 e8 ff ff ff 4f 00 19d+01:46:18.506 READ FPDMA QUEUED Error 1 occurred at disk power-on lifetime: 10463 hours (435 days + 23 hours) When the command that caused the error occurred, the device was active or idle. After command completion occurred, registers were: ER ST SC SN CL CH DH -- -- -- -- -- -- -- 40 51 00 78 95 01 0e Error: UNC at LBA = 0x0e019578 = 234984824 Commands leading to the command that caused the error were: CR FR SC SN CL CH DH DC Powered_Up_Time Command/Feature_Name -- -- -- -- -- -- -- -- ---------------- -------------------- 60 00 f8 ff ff ff 4f 00 19d+01:46:14.191 READ FPDMA QUEUED 61 00 10 ff ff ff 4f 00 19d+01:46:14.189 WRITE FPDMA QUEUED 60 00 08 ff ff ff 4f 00 19d+01:46:14.189 READ FPDMA QUEUED 60 00 00 ff ff ff 4f 00 19d+01:46:14.187 READ FPDMA QUEUED 60 00 10 ff ff ff 4f 00 19d+01:46:14.181 READ FPDMA QUEUED Any suggestions would be most appreciated!