vorel

✅ Solution I was able to solve this by removing the modifications from @SpaceInvaderOne and configure the share as a "Time Machine" share like pictured below. I have an SSD that I was using for Time Machine only. What I did was create a dedicated cache pool, but this might be a little overkill for your needs. Creating a standard share and setting the export to "Yes/Time Machine" is working for me now. Make sure you load the share up in finder before attempting to add the Time Machine disk in System Preferences (on your mac). Important note: I've had issues with this type of configuration before (on older versions of Unraid) where the time machine would run fine on the array, then it would corrupt itself a few weeks/months in. This is why I liked what @SpaceInvaderOne did (where it was an unassigned disk and not on the array). It was the most stable solution to date. I will report back if I have any issues.

Bumping this thread to see if anyone else is experiencing this issue? Any thoughts?

Thanks! There are comments on the SpaceInvaderOne's video as well saying that same thing. I wonder if @SpaceInvaderOne has any insight on this?

Problem Time Machine does not see my SMB share after applying the 6.9 update (see attached) Environment Unraid 6.9.1 (Stable) macOS Big Sur 11.1 What I have attempted It was originally configured as an unassigned disk (following Space Invader One's video). This method was working great before the 6.9 stable update that was released earlier this month I then attempted using the built-in Time Machine support by creating a cache-pool dedicated for time machine. It returned the same results. I appreciate your help! I am very grateful for the Unraid community. Let me know if there is any more detail that I can share to make it easier to ask for help.

You can try one stick at a time to see if that exposes any clues. What I learned was when I tested one stick at a time, Memtest would pass. It wasn't until I was running all four sticks is where I had an issue. I was very skeptical when ASRock told me to get new memory, but it hasn't crashed since my original post.

It ended up being a memory issue with my server. I was running 4 DIMMs. 2 of them I purchased in 2017. I purchased 2 more in 2018. It ran fine for a year until I had the issues. Strange thing was, testing each stick individually with Memtest would work. If I put all four DIMMs in there, it would fail. I contacted my motherboard manufacturer (ASRock) and they said even though I had the EXACT same part number for memory in all four slots, they said it’s because I didn’t buy the memory all together is why I had the problem. I ordered new memory and my server hasn’t crashed at all. Update your BIOS and try different memory if you have it. Good luck! Hope this helps you.

Just wanted to share another update. I just cleared all of my syslogs and it failed again within ~30 minutes. Attached is fresh syslog and picture of console. syslog.zip

Hi all, I've been troubleshooting some random freezes that I cannot replicate and I would appreciate any other advice on what to do next. Background These random freezes started happening while I was on vacation (meaning that I didn't make any changes to the system when it started happening) These freezes started on 6.7.2 I wasn't too concerned yet because I knew that 6.8.0 was around the corner and I was doing some major holiday hardware upgrades soon What I upgraded since crashes started happening I upgraded from 6.7.2 → 6.8.0 I upgraded & replaced all of my data & cache drives I upgraded my processor from AMD Ryzen 1600 to AMD Ryzen 2700X I replaced my Unraid flash drive (just because the previous one was getting old) What has stayed the same My motherboard My memory (8GB x 4 DIMMS) Tests I have run (but still have random freezes) Disabling VMs and Docker (through the options on Web GUI) MemTest (passed -- see attached) Upgraded BIOS (to latest recommended for my CPU version by ASROCK) Resetting all BIOS settings Enabled/disabled C6 states (for AMD Ryzen processors) What's strange I cannot replicate the issue This all started happening when I was not making any hardware or software changes Sometimes there are no errors at the time of the freeze, other times I get a lot of messages The RIP error values are not always the same (see my last syslog error and compare it to my actual attached console errors) The latest error that I received in my syslog is: kernel: RIP: 0010:get_page_from_freelist+0x252/0xd0b Attached are my diagnostics and my syslog (been recording it to the flash drive ever since I upgraded my processor). I would greatly appreciate any other perspectives on what to try next. Thanks in advance! syslog zeppelin-diagnostics-20191223-0842.zip

So this definitely is a wired vs wireless issue. For people who are interested why, it is quite easy for there to be some confusion. Although I have 1300 Mbps throughput, my wireless connection is "half duplex". This means only one person can talk at a time (like a walkie-talkie) I have high-end WiFi equipment, but consumer devices are still half duplex. Gigabit connection is "Full duplex". This means the throughput is slower, but is bidirectional. This makes it extremely fast compared to wireless (especially in TCP connections). Wow, today I learned a lot 😀Hope this helps! More here if you are interest in reading:

Thanks @John_M! So I tested with a WIRED Windows 10 machine and I have much better results (attached). I ordered a "USB-C -> Gigabit adapter" so I can test with my Mac (to make sure it isn't a macOS issue). So far this looks promising, but I am still confused on why wired would be faster (since my wireless is faster than Gigabit and I also have good signal). I will post my results once the new adapter arrives for my Macbook.

Hi all, I am experiencing slower than expected writes to my Cache drive. Here is my setup: Unraid Server Version: Unraid 6.5.3 Cache Drive: 1 X Crucial MX500 1TB 3D NAND SATA M.2 Network Gigabit NIC on motherboard Gigabit switch 802.11ac (with macOS reporting 1170 Mbps of Wifi "Tx Rate") My Problem When I try to transfer a file, I get about 18 MB/s (see "1-TransferRate.png" attached) Expected Results According to the Amazon data, I should expect "Sequential reads/writes up to 560/510 MB/s and random reads/writes up to 95k/90k on all file types". I was hoping to get something at least in the ~100 MB/s range My attempts so far I attempted enabling "DirectIO" by configuring "Settings -> Global Share Settings" to "Yes" I have my "Test" share set to use Cache Disk as "Only" (see "2-CacheShare" attached) I performed the transfer test with all VMs and Docker containers powered off, but I still had the same transfer results None of my attempts have seemed to improve the transfer speed. Does anyone have any thoughts?

Thanks @johnnie.black! My network transfers were also slow, so it was just a coincidence that `dd` was producing slow results as well. I continued some troubleshooting and this seemed to resolved my issue: I enabled `Direct IO` per this post Once I had that enabled, my speeds were dramatically better over the network (`dd` was still slow per your bug suggestion) Some docker containers failed to start though, so I followed these steps to delete the Docker image and pull everything back down again Looking good now, thanks! ?

Hi everyone, I searched the forums, but I could not find an answer to this. Attached are my test results for testing my disk speeds. You can see that when I write to the cache share it has a huge performance decrease. When I write directly to the drive (`/mnt/cache`) you can see my speeds are incredibly improved. I am running the disk speed test suggested from this post. I am also running Unraid 6.5.3. Any thoughts on why the share is so much slower?

I found this on a Google Search and I am experiencing the exact same issue. I am running a Late 2016 i7 MacBook Pro over AC to Unraid 6.3.5. When I view my transfer rate for my wireless, my Macbook reports that it is getting 867mbps of transfer (which that access-point then connects to my Unraid server on a gigabit connection). Moving files over SMB works fine, but just the Time Machine is slow (averaging about 1 MB/s or less). Setting my AFP Time Machine share to use my Cache disk has not helped it at all. Has anyone found any solutions to this?