GoChris Posted August 28, 2022 Share Posted August 28, 2022 Wondering if upgrading my HBA SAS 6G cards will net me more speed in rebuilds and parity checks. Case: Norco 4224, 6 backplanes, I'm told they are SAS 3 capable. Drives on all backplanes. Drives: All 7200rpm Current HBAs: Two 2-port sas2 cards. Card 1: both ports -> intel expander -> 4 ports to backplanes Card 2: both ports to remaining 2 backplanes Will upgrading to a SAS3 12Gbps card actually increase the hdd speeds at all? Cheers Quote Link to comment
JorgeB Posted August 28, 2022 Share Posted August 28, 2022 6 hours ago, GoChris said: Card 2: both ports to remaining 2 backplanes It won't for this one. 6 hours ago, GoChris said: Card 1: both ports -> intel expander -> 4 ports to backplanes Which Intel expander? Also what model disks? Quote Link to comment
GoChris Posted August 28, 2022 Author Share Posted August 28, 2022 (edited) 6 hours ago, JorgeB said: Which Intel expander? Also what model disks? Intel RES2SV240 expander Drives are a combination of: HGST Deskstar NAS Seagate Ironwolf (NAS and Pros) And just some Seagate NAS and shucked desktop drives, might have a couple SMRs in there. Not deployed yet to upgrade some 4tb: WD Gold, Toshiba X300s (desktop performace they say) Edit: I do have a 2nd expander (same model) that's unused, unsure if it would help in any way. Edited August 28, 2022 by GoChris Quote Link to comment
JorgeB Posted August 29, 2022 Share Posted August 29, 2022 16 hours ago, GoChris said: Intel RES2SV240 expander This is SAS2, so having a SAS3 HBA won't change the link speed, it will still help if your board/CPU supports PCIe 3.0, same as a SAS2 HBA with PCIe 3.0 support would, like for example the 9207-8i. Quote Link to comment
GoChris Posted August 29, 2022 Author Share Posted August 29, 2022 6 hours ago, JorgeB said: This is SAS2, so having a SAS3 HBA won't change the link speed, it will still help if your board/CPU supports PCIe 3.0, same as a SAS2 HBA with PCIe 3.0 support would, like for example the 9207-8i. Both cards currently are LSI SAS2008 based. Quote Link to comment
JorgeB Posted August 29, 2022 Share Posted August 29, 2022 Those are PCIe 2.0, a PCIe 3.0 model connected to the expander with dual link would increase total bandwidth from the currently usable 3GB/s to 4.8GB/s, assuming of course the board/CPU supports PCIe 3.0. Quote Link to comment
GoChris Posted August 29, 2022 Author Share Posted August 29, 2022 18 minutes ago, JorgeB said: Those are PCIe 2.0, a PCIe 3.0 model connected to the expander with dual link would increase total bandwidth from the currently usable 3GB/s to 4.8GB/s, assuming of course the board/CPU supports PCIe 3.0. Yes, the board does, thanks for all the info so far. So in theory if all drives on the backplanes off the expander are populated (16), currently at 3GB/s would result in a max of 192MB/s per drive? If so, doesn't seem like an upgrade would net much with current HDD speeds. Quote Link to comment
JorgeB Posted August 29, 2022 Share Posted August 29, 2022 9 minutes ago, GoChris said: currently at 3GB/s would result in a max of 192MB/s per drive? Yes, around that. Quote Link to comment
JorgeB Posted August 29, 2022 Share Posted August 29, 2022 If you are getting lower speeds than that during rebuilds/parity check HBAs are not the problem. Quote Link to comment
GoChris Posted August 29, 2022 Author Share Posted August 29, 2022 (edited) 5 minutes ago, JorgeB said: If you are getting lower speeds than that during rebuilds/parity check HBAs are not the problem. I am yes, alright I wonder what it could be. I rebuilt a 10TB a few days ago @ 116.5 MB/s. Good to know. CPU is an AMD Ryzen 7 3800X Edited August 29, 2022 by GoChris Quote Link to comment
JorgeB Posted August 29, 2022 Share Posted August 29, 2022 Check the HBAs link speed and with, you can do that with lspci -d 1000: -vv Also check that the expander is using dual link cat /sys/class/sas_host/host1/device/port-1\:0/sas_port/port-1\:0/num_phys if it's not host1 you need to adjust the command, e.g. for host7 it would be cat /sys/class/sas_host/host7/device/port-7\:0/sas_port/port-7\:0/num_phys Output of 8 means dual link, 4 means single link, you can also post the diags if you want me to check the correct host for you. Quote Link to comment
GoChris Posted August 29, 2022 Author Share Posted August 29, 2022 Quote Check the HBAs link speed and with, you can do that with 09:00.0 Serial Attached SCSI controller: Broadcom / LSI SAS2008 PCI-Express Fusion-MPT SAS-2 [Falcon] (rev 03) Subsystem: Broadcom / LSI SAS2008 PCI-Express Fusion-MPT SAS-2 [Falcon] 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 81 IOMMU group: 17 Region 0: I/O ports at e000 [size=256] Region 1: Memory at d04c0000 (64-bit, non-prefetchable) [size=16K] Region 3: Memory at d0080000 (64-bit, non-prefetchable) [size=256K] Expansion ROM at d0000000 [disabled] [size=512K] 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: CorrErr+ NonFatalErr+ FatalErr+ UnsupReq+ RlxdOrd+ ExtTag+ PhantFunc- AuxPwr- NoSnoop+ FLReset- MaxPayload 512 bytes, MaxReadReq 512 bytes DevSta: CorrErr- NonFatalErr- FatalErr- UnsupReq- AuxPwr- TransPend- LnkCap: Port #0, Speed 5GT/s, Width x8, ASPM L0s, Exit Latency L0s <64ns ClockPM- Surprise- LLActRep- BwNot- ASPMOptComp- LnkCtl: ASPM Disabled; RCB 64 bytes, Disabled- CommClk+ ExtSynch- ClockPM- AutWidDis- BWInt- AutBWInt- LnkSta: Speed 5GT/s (ok), Width x8 (ok) TrErr- Train- SlotClk+ DLActive- BWMgmt- ABWMgmt- DevCap2: Completion Timeout: Range BC, TimeoutDis+ NROPrPrP- LTR- 10BitTagComp- 10BitTagReq- OBFF Not Supported, ExtFmt- EETLPPrefix- EmergencyPowerReduction Not Supported, EmergencyPowerReductionInit- FRS- TPHComp- ExtTPHComp- AtomicOpsCap: 32bit- 64bit- 128bitCAS- DevCtl2: Completion Timeout: 50us to 50ms, TimeoutDis- LTR- OBFF Disabled, AtomicOpsCtl: ReqEn- 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- Retimer- 2Retimers- CrosslinkRes: unsupported Capabilities: [d0] Vital Product Data pcilib: sysfs_read_vpd: read failed: No such device 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=00002000 PBA: BAR=1 offset=00003800 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- AdvNonFatalErr- CEMsk: RxErr- BadTLP- BadDLLP- Rollover- Timeout- AdvNonFatalErr+ AERCap: First Error Pointer: 00, ECRCGenCap+ ECRCGenEn- ECRCChkCap+ ECRCChkEn- MultHdrRecCap- MultHdrRecEn- TLPPfxPres- HdrLogCap- HeaderLog: 00000000 00000000 00000000 00000000 Capabilities: [138 v1] Power Budgeting <?> Capabilities: [150 v1] Single Root I/O Virtualization (SR-IOV) IOVCap: Migration-, Interrupt Message Number: 000 IOVCtl: Enable- Migration- Interrupt- MSE- ARIHierarchy- IOVSta: Migration- Initial VFs: 16, Total VFs: 16, Number of VFs: 0, Function Dependency Link: 00 VF offset: 1, stride: 1, Device ID: 0072 Supported Page Size: 00000553, System Page Size: 00000001 Region 0: Memory at 00000000d04c4000 (64-bit, non-prefetchable) Region 2: Memory at 00000000d00c0000 (64-bit, non-prefetchable) VF Migration: offset: 00000000, BIR: 0 Capabilities: [190 v1] Alternative Routing-ID Interpretation (ARI) ARICap: MFVC- ACS-, Next Function: 0 ARICtl: MFVC- ACS-, Function Group: 0 Kernel driver in use: mpt3sas Kernel modules: mpt3sas 0a:00.0 Serial Attached SCSI controller: Broadcom / LSI SAS2008 PCI-Express Fusion-MPT SAS-2 [Falcon] (rev 03) Subsystem: Broadcom / LSI SAS2008 PCI-Express Fusion-MPT SAS-2 [Falcon] 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 89 IOMMU group: 18 Region 0: I/O ports at d000 [size=256] Region 1: Memory at d0ac0000 (64-bit, non-prefetchable) [size=16K] Region 3: Memory at d0680000 (64-bit, non-prefetchable) [size=256K] Expansion ROM at d0600000 [disabled] [size=512K] 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: CorrErr+ NonFatalErr+ FatalErr+ UnsupReq+ RlxdOrd+ ExtTag+ PhantFunc- AuxPwr- NoSnoop+ FLReset- MaxPayload 512 bytes, MaxReadReq 512 bytes DevSta: CorrErr- NonFatalErr- FatalErr- UnsupReq- AuxPwr- TransPend- LnkCap: Port #0, Speed 5GT/s, Width x8, ASPM L0s, Exit Latency L0s <64ns ClockPM- Surprise- LLActRep- BwNot- ASPMOptComp- LnkCtl: ASPM Disabled; RCB 64 bytes, Disabled- CommClk+ ExtSynch- ClockPM- AutWidDis- BWInt- AutBWInt- LnkSta: Speed 5GT/s (ok), Width x8 (ok) TrErr- Train- SlotClk+ DLActive- BWMgmt- ABWMgmt- DevCap2: Completion Timeout: Range BC, TimeoutDis+ NROPrPrP- LTR- 10BitTagComp- 10BitTagReq- OBFF Not Supported, ExtFmt- EETLPPrefix- EmergencyPowerReduction Not Supported, EmergencyPowerReductionInit- FRS- TPHComp- ExtTPHComp- AtomicOpsCap: 32bit- 64bit- 128bitCAS- DevCtl2: Completion Timeout: 50us to 50ms, TimeoutDis- LTR- OBFF Disabled, AtomicOpsCtl: ReqEn- 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- Retimer- 2Retimers- CrosslinkRes: unsupported Capabilities: [d0] Vital Product Data pcilib: sysfs_read_vpd: read failed: No such device 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=00002000 PBA: BAR=1 offset=00003800 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- AdvNonFatalErr- CEMsk: RxErr- BadTLP- BadDLLP- Rollover- Timeout- AdvNonFatalErr+ AERCap: First Error Pointer: 00, ECRCGenCap+ ECRCGenEn- ECRCChkCap+ ECRCChkEn- MultHdrRecCap- MultHdrRecEn- TLPPfxPres- HdrLogCap- HeaderLog: 00000000 00000000 00000000 00000000 Capabilities: [138 v1] Power Budgeting <?> Capabilities: [150 v1] Single Root I/O Virtualization (SR-IOV) IOVCap: Migration-, Interrupt Message Number: 000 IOVCtl: Enable- Migration- Interrupt- MSE- ARIHierarchy- IOVSta: Migration- Initial VFs: 16, Total VFs: 16, Number of VFs: 0, Function Dependency Link: 00 VF offset: 1, stride: 1, Device ID: 0072 Supported Page Size: 00000553, System Page Size: 00000001 Region 0: Memory at 00000000d0ac4000 (64-bit, non-prefetchable) Region 2: Memory at 00000000d06c0000 (64-bit, non-prefetchable) VF Migration: offset: 00000000, BIR: 0 Capabilities: [190 v1] Alternative Routing-ID Interpretation (ARI) ARICap: MFVC- ACS-, Next Function: 0 ARICtl: MFVC- ACS-, Function Group: 0 Kernel driver in use: mpt3sas Kernel modules: mpt3sas Quote Also check that the expander is using dual link [email protected]:~# cat /sys/class/sas_host/host11/device/port-11\:0/sas_port/port-11\:0/num_phys 8 [email protected]:~# cat /sys/class/sas_host/host12/device/port-12\:0/sas_port/port-12\:0/num_phys 1 I have also attached diagnostics. I appreciate your help immensely. tower-diagnostics-20220829-1140.zip Quote Link to comment
JorgeB Posted August 29, 2022 Share Posted August 29, 2022 6 minutes ago, GoChris said: LnkSta: Speed 5GT/s (ok), Width x8 (ok) 7 minutes ago, GoChris said: LnkSta: Speed 5GT/s (ok), Width x8 (ok) HBA links are OK, expander is also using dual link 8 minutes ago, GoChris said: [email protected]:~# cat /sys/class/sas_host/host11/device/port-11\:0/sas_port/port-11\:0/num_phys 8 Something else is the problem, is 116.5MB/s the average speed or the starting speed? i.e., if you start a parity check a what speed does it start? Make sure nothing else is using the array. Quote Link to comment
GoChris Posted August 29, 2022 Author Share Posted August 29, 2022 2 hours ago, JorgeB said: HBA links are OK, expander is also using dual link Something else is the problem, is 116.5MB/s the average speed or the starting speed? i.e., if you start a parity check a what speed does it start? Make sure nothing else is using the array. That was the avg speed from the history, so it would be faster at one point. I will do a test and report back after stopping any activity. Quote Link to comment
GoChris Posted August 29, 2022 Author Share Posted August 29, 2022 2 hours ago, JorgeB said: Something else is the problem, is 116.5MB/s the average speed or the starting speed? i.e., if you start a parity check a what speed does it start? Make sure nothing else is using the array. Alright in the first 5 minutes, it was between ~162 to 172MB/s, seemed pretty stable once it got going in the high 160s. I know drives slow down on the outer edges, so perhaps some of the older 4TB drives are quite slow? I will do some speed tests on all of them at different positions to see if anyone individually is slow. Average still seems low compared to the starting speed. I thought most dockers were stopped during the last rebuild, but I can't be certain now. Quote Link to comment
GoChris Posted August 30, 2022 Author Share Posted August 30, 2022 I did a scan at 25% intervals using the Disk Speed container. The slowest 4TB drive does indeed slow to ~80MB/s at the 4TB mark. As well as one 8TB that slows to ~88MB/s at it's end. So perhaps the slow performance of those at their edges are what's dropping the average down so much. Quote Link to comment
JorgeB Posted August 30, 2022 Share Posted August 30, 2022 4 hours ago, GoChris said: So perhaps the slow performance of those at their edges are what's dropping the average down so much. Most likely, and it's not a lack of bandwidth problem, that could be the problem if the check started say at 180/190MB/s and after a couple of hours it was still going at that speed. 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.