Joseph Posted April 17, 2017 Author Share Posted April 17, 2017 Quote I will be switching out the motherboard for a newer chipset with 3 PCIex16 slots instead of the current board which has 2, this allows me to add my 4 ports NIC. Make sure they are true x16 slots. My Mobo has 3 full length slots; the first one is x16 only if nothing is installed in #2; then both are x8 on each slot. Slot 3 only operates at x4.... No way to get 2 HBAs and a decent graphics card installed without a bottleneck somewhere. grrr... Quote Link to comment
crowdx42 Posted April 17, 2017 Share Posted April 17, 2017 Well for me then I would just want to make sure the NIC ends up in the x4 slot as it really does not need x8/x16 bandwidth I think Quote Link to comment
Joseph Posted April 17, 2017 Author Share Posted April 17, 2017 True! Now you're making me want to buy a 4x1Gb/s NIC!! Quote Link to comment
Joseph Posted April 17, 2017 Author Share Posted April 17, 2017 10 hours ago, johnnie.black said: ATA errors are gone, ACPI errors can usually be ignored, bios update may help some. Problem is that the bad cable corrupted your cache pool... Ok, I followed your instructions and was able to backup the cache drive; cleared the partition on both, add them back into the pool and restore the data. No errors now, per say, just the ACPI warnings; sadly the mobo is EOL, so I'm afraid I have the latest BIOS dated 11/09/2012. Again, SAS2LP is not installed and the on-board Marvell controller has been disabled in the BIOS. The tunables entries have been adjusted per your suggestions I am going to start the parity checks now and will post the results. Thank you again for everything! Quote Link to comment
Joseph Posted April 18, 2017 Author Share Posted April 18, 2017 (edited) UPDATE: I ran 2 parity checks, with a reboot in between each and I'm happy to report there were 0 errors on both checks! So I believe the issue has been finally resolved! YAY!! However, with the old configuration, a typical parity check took around 8 or so hours to complete @ about 125MB/s Now the parity checks are taking almost 12 hours at about 80MB/s or 90MB/s. I noticed a speed drop using MC too. Could it be that the tunables entries need to be adjusted? Or is it that it will be slower now that the parity drives are on the Mobo SATA II controller? Would Turbo mode be worth a consideration? Tunable (enable NCQ): No Tunable (nr_requests): 128 default Tunable (md_num_stripes): 4096 user-set Tunable (md_sync_window): 2048 user-set Tunable (md_sync_thresh): 2000 user-set Edited April 18, 2017 by Joseph Quote Link to comment
JorgeB Posted April 18, 2017 Share Posted April 18, 2017 5 minutes ago, Joseph said: Could it be that the tunables entries need to be adjusted? Or is it that it will be slower now that the parity drives are on the Mobo SATA II controller? Would Turbo mode be worth a consideration? Tunables I posted are not optimized for your setup, but they are usually good for most configs, you should run the tunables tester when available for >v6.2. SATA2 is enough for current HDDs, turbo mode will help with writing speed, not parity checks. LSI usually performs same or better than the SAS2LP, strange yours slowed down considerably, is the controller in the same slot as the old one? is it on a CPU slot (one of the top 2 slots)? Quote Link to comment
Joseph Posted April 18, 2017 Author Share Posted April 18, 2017 On 4/12/2017 at 3:06 PM, Joseph said: forgot to mention right before shutdown this last time, I noticed a line that flashed on the screen quickly that I think said ACPI error. What effect on unRAID would it have? How can errors on shutdown be captured for analysis? re: my post from last week, here's the actual shutdown error I have. I think it has something to do with the UPS. Anyone have thoughts on how to fix? Quote Link to comment
Joseph Posted April 18, 2017 Author Share Posted April 18, 2017 (edited) 6 minutes ago, johnnie.black said: Tunables I posted are not optimized for your setup, but they are usually good for most configs, you should run the tunables tester when available for >v6.2. SATA2 is enough for current HDDs, turbo mode will help with writing speed, not parity checks. LSI usually performs same or better than the SAS2LP, strange yours slowed down considerably, is the controller in the same slot as the old one? is it on a CPU slot (one of the top 2 slots)? Currently it is in Slot #2. I only have 3 slots, Slot #1, I have an el-cheapo gpu, but plan to use for a better one for VMs soon. Slot #2 is the LSI. In this config, the lanes are on 1 & 2 are 8x. Slot #3, is a 4x lane only slot. I am going to have to buy another H310 so I can have all drive cages connected. Not sure the best way to have all 3 installed. Edited April 18, 2017 by Joseph Quote Link to comment
JorgeB Posted April 18, 2017 Share Posted April 18, 2017 3 hours ago, Joseph said: Slot #3, is a 4x lane only slot. I am going to have to buy another H310 so I can have all drive cages connected. Not sure the best way to have all 3 installed. Best would be the other CPU slot, but if you're using it for a GPU you'll need to used the bottom slot, not ideal not only because it's x4 but mostly because it will share the usable bandwidth, about 1500MB/s, between that controller and the onboard SATA ports. Post the output of: lspci -vv -d 1000:* To confirm the LSI is linking at the correct speed. Quote Link to comment
Joseph Posted April 18, 2017 Author Share Posted April 18, 2017 (edited) 12 minutes ago, johnnie.black said: Post the output of: lspci -vv -d 1000:* Here it is... I see an I/O error. Thoughts? Quote 02:00.0 Serial Attached SCSI controller: LSI Logic / Symbios Logic SAS2008 PCI-Express Fusion-MPT SAS-2 [Falcon] (rev 03) Subsystem: Dell 6Gbps SAS HBA Adapter Control: I/O+ Mem+ BusMaster+ SpecCycle- MemWINV- VGASnoop- ParErr- Stepping- SERR- FastB2B- DisINTx+ Status: Cap+ 66MHz- UDF- FastB2B- ParErr- DEVSEL=fast >TAbort- <TAbort- <MAbort- >SERR- <PERR- INTx- Latency: 0, Cache Line Size: 64 bytes Interrupt: pin A routed to IRQ 17 Region 0: I/O ports at e000 Region 1: Memory at f6e40000 (64-bit, non-prefetchable) Region 3: Memory at f6e00000 (64-bit, non-prefetchable) Expansion ROM at f6d00000 [disabled] Capabilities: [50] Power Management version 3 Flags: PMEClk- DSI- D1+ D2+ AuxCurrent=0mA PME(D0-,D1-,D2-,D3hot-,D3cold-) Status: D0 NoSoftRst+ PME-Enable- DSel=0 DScale=0 PME- Capabilities: [68] Express (v2) Endpoint, MSI 00 DevCap: MaxPayload 4096 bytes, PhantFunc 0, Latency L0s <64ns, L1 <1us ExtTag+ AttnBtn- AttnInd- PwrInd- RBE+ FLReset+ SlotPowerLimit 0.000W DevCtl: Report errors: Correctable+ Non-Fatal+ Fatal+ Unsupported+ RlxdOrd- ExtTag- PhantFunc- AuxPwr- NoSnoop+ FLReset- MaxPayload 128 bytes, MaxReadReq 512 bytes DevSta: CorrErr+ UncorrErr- FatalErr- UnsuppReq+ AuxPwr- TransPend- LnkCap: Port #0, Speed 5GT/s, Width x8, ASPM L0s, Exit Latency L0s <64ns, L1 <1us ClockPM- Surprise- LLActRep- BwNot- ASPMOptComp- LnkCtl: ASPM Disabled; RCB 64 bytes Disabled- CommClk+ ExtSynch- ClockPM- AutWidDis- BWInt- AutBWInt- LnkSta: Speed 5GT/s, Width x8, TrErr- Train- SlotClk+ DLActive- BWMgmt- ABWMgmt- DevCap2: Completion Timeout: Range BC, TimeoutDis+, LTR-, OBFF Not Supported DevCtl2: Completion Timeout: 50us to 50ms, TimeoutDis-, LTR-, OBFF Disabled LnkCtl2: Target Link Speed: 5GT/s, EnterCompliance- SpeedDis- Transmit Margin: Normal Operating Range, EnterModifiedCompliance- ComplianceSOS- Compliance De-emphasis: -6dB LnkSta2: Current De-emphasis Level: -6dB, EqualizationComplete-, EqualizationPhase1- EqualizationPhase2-, EqualizationPhase3-, LinkEqualizationRequest- Capabilities: [d0] Vital Product Datapcilib: sysfs_read_vpd: read failed: Input/output error Not readable Capabilities: [a8] MSI: Enable- Count=1/1 Maskable- 64bit+ Address: 0000000000000000 Data: 0000 Capabilities: [c0] MSI-X: Enable+ Count=15 Masked- Vector table: BAR=1 offset=0000e000 PBA: BAR=1 offset=0000f800 Capabilities: [100 v1] Advanced Error Reporting UESta: DLP- SDES- TLP- FCP- CmpltTO- CmpltAbrt- UnxCmplt- RxOF- MalfTLP- ECRC- UnsupReq- ACSViol- UEMsk: DLP- SDES- TLP- FCP- CmpltTO- CmpltAbrt- UnxCmplt- RxOF- MalfTLP- ECRC- UnsupReq- ACSViol- UESvrt: DLP+ SDES+ TLP- FCP+ CmpltTO- CmpltAbrt- UnxCmplt- RxOF+ MalfTLP+ ECRC- UnsupReq- ACSViol- CESta: RxErr- BadTLP- BadDLLP- Rollover- Timeout- NonFatalErr- CEMsk: RxErr- BadTLP- BadDLLP- Rollover- Timeout- NonFatalErr+ AERCap: First Error Pointer: 00, GenCap+ CGenEn- ChkCap+ ChkEn- Capabilities: [138 v1] Power Budgeting <?> Kernel driver in use: mpt3sas Kernel modules: mpt3sas Edited April 18, 2017 by Joseph Quote Link to comment
JorgeB Posted April 18, 2017 Share Posted April 18, 2017 3 minutes ago, Joseph said: I see an I/O error. Thoughts? That's normal, and link speed is correct, PCIe 2.0 x8, so maybe the tunables need optimization, keep an eye on this thread for the new version: Quote Link to comment
Joseph Posted April 19, 2017 Author Share Posted April 19, 2017 (edited) 16 hours ago, johnnie.black said: Best would be the other CPU slot, but if you're using it for a GPU you'll need to used the bottom slot, not ideal not only because it's x4 but mostly because it will share the usable bandwidth, about 1500MB/s, between that controller and the onboard SATA ports. What are your thoughts of not using any of the on-board SATA ports and use this configuration: Slot 1 (x8 speed): Dell H310 #1 Slot 2 (x8 speed): Dell H310 #2 Slot 4 (x4 speed): GPU <~~~ I suppose this means the Nvidia GTX 1060 is out of the question due to physical size (space constraints) and limited lanes? Is there a card that would work for VM gaming? Edited April 19, 2017 by Joseph Quote Link to comment
crowdx42 Posted April 19, 2017 Share Posted April 19, 2017 One question that might be relevant in this thread, SATA cable length? I am thinking that 3ft would allow for better cable management in a rack chassis but will it affect the connection. What lengths are you guys using? Quote Link to comment
Joseph Posted April 19, 2017 Author Share Posted April 19, 2017 1 hour ago, crowdx42 said: What lengths are you guys using? My longest run to the mobo is 2ft and the SFF cables to the HBA are about 2.47ft. I somewhere that the max spec is 1 meter so wouldn't advise going over 3ft. Quote Link to comment
JorgeB Posted April 19, 2017 Share Posted April 19, 2017 3 hours ago, Joseph said: What are your thoughts of not using any of the on-board SATA ports and use this configuration: Slot 1 (x8 speed): Dell H310 #1 Slot 2 (x8 speed): Dell H310 #2 Slot 4 (x4 speed): GPU <~~~ I suppose this means the Nvidia GTX 1060 is out of the question due to physical size (space constraints) and limited lanes? Is there a card that would work for VM gaming? No real experience with that so difficult to offer advice. Quote Link to comment
Recommended Posts
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.