January 5, 201412 yr I just recently built a new setup with some old drives I had setup 18 months ago, and seem to be having some serious transfer speed issues – I’m getting 200 KB/s. I also seem to be hitting access problems trying to transfer things after 30-40 seconds. Setup version: 5.0.4 Basic Asus H87I-Plus G.Skill Sniper 8GB (2x 4GB) DDR3-1600 Lian-Li PC-Q25B Silverstone 450W SFX12V Intel Core i3-4130 3.4GHz Dual-Core Processor Seagate Barracuda – 2TB - Parity Seagate Barracuda – 2TB Seagate Barracuda – 1TB I ran smartctrl on all of my drives and didn’t see anything weird except for one drive which is coming back with an odd Command Timeout value as well as a UDMA_CRC_Error_Count above 0. I tried swapping cabling and positions for that drive and I’m still seeing the problem. I am also kind of a neophyte when it comes to Unraid, so I might be missing something glaringly obvious. I’ve tried troubleshooting but am stumped at this point. Vendor Specific SMART Attributes with Thresholds: ID# ATTRIBUTE_NAME FLAG VALUE WORST THRESH TYPE UPDATED WHEN_FAILED RAW_VALUE 1 Raw_Read_Error_Rate 0x000f 119 099 006 Pre-fail Always - 218810564 3 Spin_Up_Time 0x0003 100 100 000 Pre-fail Always - 0 4 Start_Stop_Count 0x0032 100 100 020 Old_age Always - 205 5 Reallocated_Sector_Ct 0x0033 100 100 036 Pre-fail Always - 28 7 Seek_Error_Rate 0x000f 063 060 030 Pre-fail Always - 2266956 9 Power_On_Hours 0x0032 091 091 000 Old_age Always - 7887 10 Spin_Retry_Count 0x0013 100 100 097 Pre-fail Always - 0 12 Power_Cycle_Count 0x0032 100 100 020 Old_age Always - 202 183 Runtime_Bad_Block 0x0032 097 097 000 Old_age Always - 3 184 End-to-End_Error 0x0032 100 100 099 Old_age Always - 0 187 Reported_Uncorrect 0x0032 100 100 000 Old_age Always - 0 188 Command_Timeout 0x0032 099 099 000 Old_age Always - 12885098499 189 High_Fly_Writes 0x003a 097 097 000 Old_age Always - 3 190 Airflow_Temperature_Cel 0x0022 063 049 045 Old_age Always - 37 (Min/Max 35/40) 194 Temperature_Celsius 0x0022 037 051 000 Old_age Always - 37 (0 18 0 0 0) 195 Hardware_ECC_Recovered 0x001a 041 020 000 Old_age Always - 218810564 197 Current_Pending_Sector 0x0012 100 100 000 Old_age Always - 0 198 Offline_Uncorrectable 0x0010 100 100 000 Old_age Offline - 0 199 UDMA_CRC_Error_Count 0x003e 200 200 000 Old_age Always - 12 240 Head_Flying_Hours 0x0000 100 253 000 Old_age Offline - 212437672403129 241 Total_LBAs_Written 0x0000 100 253 000 Old_age Offline - 1398830540 242 Total_LBAs_Read 0x0000 100 253 000 Old_age Offline - 3620789049 Relevant smartctl Log and Sys log attached. Sys_Log.txt issue_smartctl.txt
January 6, 201412 yr The motherboard/CPU/memory and power supply are all excellent choices and shouldn't have any issues with this setup. I'd do a couple of things to help isolate what's going wrong here ... (a) Run MemTest for at least 2 hours and confirm it doesn't find any issues with your memory. (b) Be certain your Ethernet cables are firmly seated on at all connections (on the UnRAID server; at the router/switch; at your PC; etc.). In fact, I'd try DIFFERENT cables if possible. © Do a "New Config" (on the Utils tab) and only assign ONE data drive ... nothing else. Then try the array and see how it works. If it still has the slow speed/access issue, redo the New Config ... again with a single data disk (but a different one this time) (d) If you're still having issues, create a new UnRAID Flash drive just to confirm that you don't have an issue with that. Use a different flash drive ... and build it from scratch with the v5.04 download.
January 6, 201412 yr Author Thanks for the advice, A B and C didn't seem to resolve anything or reveal additional problems (all drives had transfer speed issues, no issues with the MemTest). I'm going to pick up a new flash drive and try a fresh install of UnRAID. I also just realized I cut the MemTest off a bit shorter than you recommended, so I'll let that run longer just in case. As an aside, parity checks seem to be hitting about 130 MB/s, not sure if that gives any additional indications of where the problem might lie.
January 6, 201412 yr Parity checks at 130MB/s indicates the system is running VERY well ... remember that ALL drives are involved in this, and they're obviously all running fine. Note that even the drive you listed the SMART values for isn't "failed" ... it's got a couple of parameters that are less than optimum, but still looks fine (and clearly is working fine). This is almost certainly a network issue -- you may have a bad Ethernet cable; a bad port on your router (or switch if there's a switch involved); or possible a bad adapter at one end. Try both a different cable AND a different port on your router/switch. Are there other PC's on the network besides just your main PC and the UnRAID server? If so, do a transfer between two of the other PCs and see what kind of speed you get; and if it looks okay, then use the SAME two ports those PC's were using to test your main PC with the UnRAID server.
January 6, 201412 yr Note: You could also use a direct connection between the UnRAID server and a PC ... no router or switch involved ... if you assign a static IP to both UnRAID and the PC.
January 7, 201412 yr Author ifconfig ethtool eth0 Looks like I'm getting a ton of dropped packets. I did a direct connection and now I'm getting 10-11 MB/s transfer speeds (hindered by my 100 Mb nic). Seems like my router is screwing me over in some way.
January 7, 201412 yr I'm not at all surprised. As I noted above, this HAD to be a networking issue, given the other data you had provided. You're getting full 100Mb/s transfers with the direct connection ... so you've either got a bad port (or ports) on your router; or you've got a bad cable somewhere in the mix. The good news is that's pretty easy to isolate ... you just need to do a bit of experimentation with cables and/or another router. [if you happen to have a switch in your topology, be sure to confirm that it's not the issue]
January 7, 201412 yr Author I've concluded that it is the router itself - it needed to be replaced anyway. Thanks for all the help, glad that it was just network issues in the end.
Archived
This topic is now archived and is closed to further replies.