September 18, 20223 yr Hello, I have started to notice that the transfer speed at time we very slow as in the KB/s and I would get a pop mesg when opening a fold from another computer. telling me to wait for it to open or cancel it and this waitr could take mins. So I went to upgrade the system with new hard drives and files system. on the transfer from the old hard drive at times it would be really 180MB/s but then it would take longer time like all days for some folders with smaller files in them. I would see the copying file stuck for mins, then the rest of the files fly by and them stops again for 10 mins or more on a 3 meg file. I dont know if the old hard drive are slow because it's almost full got like 3 - 30 gb left on them. What I notice is that trying to access the new hard drives with the new file system is just as bad. I cant copy files across the network it would time out and streaming now working. Lots of room on the new hard drives and computer is a beast. So if anyone has some ideas that will help me solve these slow proformance please let me know what I can do. p.s. after i move all my data half the hard drive and ssd will be removed kraken-diagnostics-20220918-1356.zip
September 19, 20223 yr I see several things repeating in the log that do not seem normal : Sep 18 13:48:13 KRAKEN zed[6752]: Diagnosis Engine: error event 'ereport.fs.zfs.deadman' Sep 18 13:48:13 KRAKEN zed[6752]: Missed 25 events Sep 18 13:48:13 KRAKEN zed[6752]: Bumping queue length to 2147483647 Sep 18 13:48:13 KRAKEN zed[6752]: Diagnosis Engine: error event 'ereport.fs.zfs.deadman' Sep 18 13:48:13 KRAKEN zed[6752]: Missed 20 events Sep 18 13:48:13 KRAKEN zed[6752]: Bumping queue length to 2147483647 Sep 18 13:48:13 KRAKEN zed[6752]: Diagnosis Engine: error event 'ereport.fs.zfs.deadman' Sep 18 13:48:22 KRAKEN kernel: sd 1:2:12:0: [sdg] tag#7259 task abort SUCCESS!! scmd(0x000000008d67e782) Sep 18 13:48:22 KRAKEN kernel: sd 1:2:12:0: [sdg] tag#7259 CDB: opcode=0x0 00 00 00 00 00 00 Sep 18 13:48:22 KRAKEN kernel: sd 1:2:12:0: [sdg] tag#7259 OCR is requested due to IO timeout!! Sep 18 13:48:22 KRAKEN kernel: sd 1:2:12:0: [sdg] tag#7259 SCSI host state: 5 SCSI host busy: 1 FW outstanding: 0 Sep 18 13:48:22 KRAKEN kernel: sd 1:2:12:0: [sdg] tag#7259 scmd: (0x000000008d67e782) retries: 0x0 allowed: 0x5 Sep 18 13:48:22 KRAKEN kernel: sd 1:2:12:0: [sdg] tag#7259 CDB: opcode=0x8a 8a 00 00 00 00 05 e6 f5 d8 60 00 00 00 b0 00 00 Sep 18 13:48:22 KRAKEN kernel: megaraid_sas 0000:13:00.0: megasas_disable_intr_fusion is called outbound_intr_mask:0x40000009 Sep 18 13:48:22 KRAKEN kernel: megaraid_sas 0000:13:00.0: megasas_enable_intr_fusion is called outbound_intr_mask:0x40000000 Sep 18 13:48:32 KRAKEN kernel: sd 1:2:12:0: Power-on or device reset occurred The first one seem to be related to ZFS, so I would suggest asking in the support thread for the plugin. The second seem to be sdg being reset. I'd guess that this device is part of a ZFS file system ? You should check the drive connections (data and power). Note that RAID controllers are not recommended even if that's maybe not the cause of your issue.
September 19, 20223 yr Author 53 minutes ago, ChatNoir said: I see several things repeating in the log that do not seem normal : Sep 18 13:48:13 KRAKEN zed[6752]: Diagnosis Engine: error event 'ereport.fs.zfs.deadman' Sep 18 13:48:13 KRAKEN zed[6752]: Missed 25 events Sep 18 13:48:13 KRAKEN zed[6752]: Bumping queue length to 2147483647 Sep 18 13:48:13 KRAKEN zed[6752]: Diagnosis Engine: error event 'ereport.fs.zfs.deadman' Sep 18 13:48:13 KRAKEN zed[6752]: Missed 20 events Sep 18 13:48:13 KRAKEN zed[6752]: Bumping queue length to 2147483647 Sep 18 13:48:13 KRAKEN zed[6752]: Diagnosis Engine: error event 'ereport.fs.zfs.deadman' Sep 18 13:48:22 KRAKEN kernel: sd 1:2:12:0: [sdg] tag#7259 task abort SUCCESS!! scmd(0x000000008d67e782) Sep 18 13:48:22 KRAKEN kernel: sd 1:2:12:0: [sdg] tag#7259 CDB: opcode=0x0 00 00 00 00 00 00 Sep 18 13:48:22 KRAKEN kernel: sd 1:2:12:0: [sdg] tag#7259 OCR is requested due to IO timeout!! Sep 18 13:48:22 KRAKEN kernel: sd 1:2:12:0: [sdg] tag#7259 SCSI host state: 5 SCSI host busy: 1 FW outstanding: 0 Sep 18 13:48:22 KRAKEN kernel: sd 1:2:12:0: [sdg] tag#7259 scmd: (0x000000008d67e782) retries: 0x0 allowed: 0x5 Sep 18 13:48:22 KRAKEN kernel: sd 1:2:12:0: [sdg] tag#7259 CDB: opcode=0x8a 8a 00 00 00 00 05 e6 f5 d8 60 00 00 00 b0 00 00 Sep 18 13:48:22 KRAKEN kernel: megaraid_sas 0000:13:00.0: megasas_disable_intr_fusion is called outbound_intr_mask:0x40000009 Sep 18 13:48:22 KRAKEN kernel: megaraid_sas 0000:13:00.0: megasas_enable_intr_fusion is called outbound_intr_mask:0x40000000 Sep 18 13:48:32 KRAKEN kernel: sd 1:2:12:0: Power-on or device reset occurred The first one seem to be related to ZFS, so I would suggest asking in the support thread for the plugin. The second seem to be sdg being reset. I'd guess that this device is part of a ZFS file system ? You should check the drive connections (data and power). Note that RAID controllers are not recommended even if that's maybe not the cause of your issue. the connection data and power are ok, unless it's in the backplan. the slow speed is happening on both the controller and on the motherbard sata ports. I dont have a clue why the transfer speed would run slow in KB/s when I have a big pipe for it to at high speed. I have shut it down because I have to replace one of the new drives and dont want my data to get lost on that mirror. So as soon as I get the new drives I will test the speed again.
Join the conversation
You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.