September 7, 20232 yr I've got a speed problem I just can't seem to track down. I first noticed it when I was no longer able to download at 100MB/s from newsgroups with SAB. (Which I've been doing for years, gig fiber up and down) The speeds we're now running around 60 to 80MB/s. I honestly chocked it up to my new router and some setting I had misconfigured and ignored it for about a month before I started poking into it. I've now discovered that for some reason transfers to the Unraid server, whether from Windows SMB or a docker like SAB aren't running at full speed to the cache drive or Unassigned Drives. In the past when transferring to the server I could watch the speed on the "MAIN" page, a constant 100MB/s give or take (1Gb network). Now however it appears to be caching somewhere before doing a burst write at around 400MB/s, then filling whatever cache it's using and bursting again, over and over until the transfer is finished. Transfers from Windows (various machines) all hover around 50 to 70MB/s. Copying FROM the cache drive does produce the results I'm use to seeing. Full network speed, and the "MAIN" page shows the constant read of around 100MB/s per usual. I've also noticed the same behavior with my Unassigned Drives as well. I have two SSD's mounted with UD, and they act EXACTLY the same as above. I want to say this all happened when I upgraded to 6.12 but honestly I can't say for sure. I waited until .3 was released before upgrading, and just recently upgraded to .4 I was previously on 6.11.5. Troubleshooting I've done: Tested multiple shares, thought admittedly not every single one. All of them are set to Cache drive Primary, and Array secondary with the mover running at night around..... 4am I think. Disabled secondary storage on one of the shares with Cache as primary. Same results Created brand new shares, one setup as usual with Cache as primary, Array secondary, Same results. Created a second share using array only, and transferring to that share shows exactly what I would expect from the MAIN page, constant writes on the data drive and parity, none of this burst writing. I'd recently disabled SMB 1.0, but I went ahead and re-enabled it just for testing. Same results I had the CA Auto Turbo Write Mode plugin installed, I've since removed it and played with the settings manually. Trying all 3 (Auto, Read/Modify/Write, and Reconstruct Write) none of which made a difference, nor would I expect it to on cache drive only writes. I've run read/write tests on the cache drive itself, it's acting perfectly normal: Read Speed hdparm -tT /dev/sdn /dev/sdn: Timing cached reads: 29548 MB in 1.99 seconds = 14876.03 MB/sec Timing buffered disk reads: 1174 MB in 3.00 seconds = 390.73 MB/sec Write speed /mnt/cache# dd if=/dev/zero of=test bs=1G count=1 oflag=direct 1+0 records in 1+0 records out 1073741824 bytes (1.1 GB, 1.0 GiB) copied, 2.78236 s, 386 MB/s Ran network tests to make sure it wasn't the network, though given it's happening on multiple machines i knew it wasn't, but i also knew i'd be asked! From Client to Unraid [ ID] Interval Transfer Bandwidth [ 4] 0.00-1.00 sec 113 MBytes 944 Mbits/sec [ 4] 1.00-2.00 sec 112 MBytes 943 Mbits/sec [ 4] 2.00-3.00 sec 112 MBytes 942 Mbits/sec [ 4] 3.00-4.00 sec 112 MBytes 944 Mbits/sec [ 4] 4.00-5.00 sec 112 MBytes 944 Mbits/sec [ 4] 5.00-6.00 sec 112 MBytes 938 Mbits/sec [ 4] 6.00-7.00 sec 109 MBytes 911 Mbits/sec [ 4] 7.00-8.00 sec 112 MBytes 936 Mbits/sec [ 4] 8.00-9.00 sec 112 MBytes 942 Mbits/sec [ 4] 9.00-10.00 sec 112 MBytes 944 Mbits/sec - - - - - - - - - - - - - - - - - - - - - - - - - [ ID] Interval Transfer Bandwidth [ 4] 0.00-10.00 sec 1.09 GBytes 939 Mbits/sec sender [ 4] 0.00-10.00 sec 1.09 GBytes 939 Mbits/sec receiver From Unraid to Client [ ID] Interval Transfer Bitrate Retr Cwnd [ 5] 0.00-1.00 sec 114 MBytes 957 Mbits/sec 0 308 KBytes [ 5] 1.00-2.00 sec 112 MBytes 942 Mbits/sec 0 302 KBytes [ 5] 2.00-3.00 sec 112 MBytes 936 Mbits/sec 0 302 KBytes [ 5] 3.00-4.00 sec 112 MBytes 938 Mbits/sec 0 297 KBytes [ 5] 4.00-5.00 sec 112 MBytes 938 Mbits/sec 0 314 KBytes [ 5] 5.00-6.00 sec 111 MBytes 934 Mbits/sec 0 311 KBytes [ 5] 6.00-7.00 sec 106 MBytes 887 Mbits/sec 0 299 KBytes [ 5] 7.00-8.00 sec 111 MBytes 934 Mbits/sec 0 319 KBytes [ 5] 8.00-9.00 sec 111 MBytes 928 Mbits/sec 0 297 KBytes [ 5] 9.00-10.00 sec 112 MBytes 938 Mbits/sec 0 299 KBytes - - - - - - - - - - - - - - - - - - - - - - - - - [ ID] Interval Transfer Bitrate Retr [ 5] 0.00-10.00 sec 1.09 GBytes 933 Mbits/sec 0 sender [ 5] 0.00-10.00 sec 1.08 GBytes 931 Mbits/sec receiver I saw another post about something similar if not exactly like this issue back in Dec 2022, but it didn't appear to have a resolution. So instead of necro'ing a post, I figured I'd make my own. The previous one I found was: Grabbed a diag file moments after a test send finished, hopefully you guys see something I don't. Thank you guys as always! storage51-diagnostics-20230906-1902.zip
September 9, 20232 yr Author I've done a bit more testing, but I still can't track down what is happening. Went ahead and booted the server into Safe Mode, exact same issue, so it's definitely not a plugin, docker, or whatever. I attempted to monitor disk activity with iostat -d, but it's showing exactly the same thing. No activity shows on any drive until it decides to burst write whatever data has been cached, then back to nothing until the next burst write. I monitored the systems RAM using HTOP hoping maybe it was getting cached there, but ram usage barely moves at all during a transfer, so thats not it. I checked the DRAM cache on the SSD I'm using which is a Samsung 860 btw. The 1 TB model has 1GB of onboard cache. My thought was that maybe the drive IS broken, and is filling it's onboard cache before attempting to write. However after multiple tests it appears the burst write starts after about 2 to 2.4GB of data has been copied. Which far exceeds the drives onboard cache. I'm at a complete loss how or why this is happening. Unless anyone has any other ideas, I guess I'll need to look into replacing the cache drive as a next step, despite it not showing any signs anything is wrong. Void4ever
September 9, 20232 yr I am experiencing somewhat a similar issue, I cant even write around 1mbps to the server, I have checked everything I can. Even when it was working I was getting like a max of 6/7mbps. Its totally pathetic. I was getting 6/7 before last update but seemingly since latest update I cant crack 1mb, whether its a coincidence or not I am unsure but its frying my head.
September 18, 20232 yr Author So just another update. I got a new drive to use for cache, however it did not resolve the issue. I didn't really expect it to because the 2 unassigned device SSD's i have on the system are showing the same issue as i mentioned above. But at least i have a larger cache drive now I went ahead and cleared the bios/cmos as a test, but again no change after setting everything up again. One thing i did realize as i was poking around is the cache drive and the other 2 SSD's i have on the system as unassigned devices are all connected to the same chipset The motherboard i'm running is an ASrock Z87 Extreme6. This board has 10 SATA ports, 6 are controlled by the intel chipset, and 4 are controlled by a ASMedia ASM1061. The cache drive and the other 2 drives are on the ASMedia controller. Can anyone tell me if the driver for that controller changed from 6.11 to 6.12? Void4ever
September 18, 20232 yr Community Expert 3 hours ago, void4ever said: Can anyone tell me if the driver for that controller changed from 6.11 to 6.12? It uses the standard AHCI driver, there shouldn't be any major changes.
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.