April 4, 20179 yr 10 minutes ago, HellDiverUK said: I'm seeing a good 3 minutes for mine, previously it was maybe 10-15 seconds. There's a definite slow-down in unpacking the bzroot images. I tried a fresh install on a different USB on my i5-7600K machine and it was pretty slow too (considering it's overclocked to 5.2GHz!). Just checked one of my servers, still less than 10 seconds, Celeron G550.
April 4, 20179 yr Something's borked then, because there's no way a FX8350 should be taking 3 minutes. The 6700K takes about 2 minutes.
April 4, 20179 yr Just updated my main server, without issues. Great job on making the gui smoother and faster. Sent from my iPhone using Tapatalk
April 7, 20179 yr On 4/2/2017 at 7:27 PM, caseyparsons said: Loaded and it boots, but now it takes around 30 minutes to load bzimage and bzroot. Was running fine before upgrade. I'm seeing something (maybe) similar, but I gave up on it after about 5 minutes assuming it was frozen. However, I can boot into GUI mode and everything works fine, boot takes maybe a minute or so. I had similar problems with 6.3.2: the GUI mode booted up with no problems but everything would hang when I tried to boot into headless mode. For 6.3.2 I was able to fix this problem via a firmware upgrade of my ASRock motherboard. For 6.3.3 I'm again unable to boot headless - the terminal prints "Loading bzImage... ok" and then it seems to freeze up, no response from the NIC or the keyboard (even ctrl-alt-delete). [EDIT, 6 months later: I upgraded to unRaid 6.5.2 and this problem went away!] Edited June 10, 20188 yr by thither
April 8, 20179 yr On April 1, 2017 at 6:03 PM, cybrnook said: I seem to suffering the finewine syndrome as well. Mine seem to only be getting faster..... As this is now not an isolated case and with irrefutable evidence in what is clearly a major breech in regression 101, Limetech has no option but to issue a formal apology and dare I say it due to such severity a free beer to their loyal customers, who sadly through no fault of their own, allowed such draconian and downright shady business practices! We will tolerate it no more! In your case, parity getting faster....disgusting!
April 9, 20179 yr What's the best way to downgrade back to 6.3.2 if 6.3.3 messes your system up? I haven't upgraded yet, but some of the issues reported have me spooked.
April 9, 20179 yr 2 minutes ago, fitbrit said: What's the best way to downgrade back to 6.3.2 if 6.3.3 messes your system up? I haven't upgraded yet, but some of the issues reported have me spooked. On the flash drive there is a folder called "previous" copy the files within it back to the root of the flash and then reboot. Side note: Personally, I think 6.3.3 is rock solid.
April 10, 20179 yr 3 hours ago, SCSI said: No issues to report. Successfully updated here. Seems that is the case here so far too.
April 11, 20179 yr No longer detects my 10G network adapter. Mellanox ConnectX-2. Updated from 6.3.2. Edited April 11, 20179 yr by tyrindor
April 11, 20179 yr No longer detects my 10G network adapter. Mellanox ConnectX-2. Updated from 6.3.2.What does dmesg say?Sent from my iPhone using Tapatalk
April 11, 20179 yr Fixed it by restarting the server again after the update was complete. Weird that it didn't detect it on the first 6.3.3 boot. Had me freaking out that my server was booting at all, until I tried the 1G network. Edited April 11, 20179 yr by tyrindor
April 16, 20179 yr Has the intel onboard pci passthrough issue been resolved? I had to downgrade back to 6.2.4. Whenever I upgraded I was unable to pass through onboard sound and I mainly use my unraid as backup/plex/Steam streamer(Through Windows 10). I've been hesitant to upgrade since it took so long for me to figure out it was the version of Unraid that was giving me issues. My current rig for unraid is as follows: System Overview unRAID system: unRAID server Plus, version 6.2.4 Model: Custom Motherboard: ASUSTeK COMPUTER INC. - PRIME Z270-AR Processor: Intel® Core™ i5-7600 CPU @ 3.50GHz HVM: Enabled IOMMU: Enabled Cache: L1 Cache = 256 kB (max. capacity 256 kB) L2 Cache = 1024 kB (max. capacity 1024 kB) L3 Cache = 6144 kB (max. capacity 6144 kB) Memory: 32 GB (max. installable capacity 64 GB) BANK 0 = 16384 MB, 2133 MHz BANK 2 = 16384 MB, 2133 MHz Network: bond0: fault-tolerance (active-backup), mtu 1500 bond0: fault-tolerance (active-backup), mtu 1500 eth0: 10000Mb/s, full duplex, mtu 1500 Kernel: Linux 4.4.30-unRAID x86_64 OpenSSL: 1.0.2j P + Q algorithm: avx (47208 MB/s) + avx2x4 (30785 MB/s) Uptime: 7 days, 9 hours, 3 minutes, 39 seconds
April 16, 20179 yr Serious bug found... Everytime I manually click "move cache" it clears all my user shares. This moved over my cache only share, and ignored my splitting preference when moving over a ton of files, moving them all to the wrong disks and breaking almost all my addons! I cannot re-add shares as it just says "share is deleted" when I try to. I repeated this progress twice to confirm. Did not happen on 6.3.2. I have 16GB of RAM (10% used), and seeing repeated Apr 16 18:17:26 UNRAID kernel: [<ffffffff810c83bf>] out_of_memory+0x3aa/0x3e5 log.txt Edited April 16, 20179 yr by tyrindor
April 16, 20179 yr 12 minutes ago, tyrindor said: Serious bug found... Everytime I manually click "move cache" it clears all my user shares. This moved over my cache only share, and ignored my splitting preference when moving over a ton of files, moving them all to the wrong disks and breaking almost all my addons! I cannot re-add shares as it just says "share is deleted" when I try to. I repeated this progress twice to confirm. Did not happen on 6.3.2. I have 16GB of RAM and seeing repeated Apr 16 18:17:26 UNRAID kernel: [<ffffffff810c83bf>] out_of_memory+0x3aa/0x3e5 log.txt You really need to post diagnostics... Log's by themselves don't tell the whole story.
April 16, 20179 yr What do you mean by diagnostics? How do I stop the mover that's in progress? It has 900GB queued up and due to not having shares it's moving the data all wrong...
April 16, 20179 yr Just now, tyrindor said: What do you mean by diagnostics? How do I stop the mover that's in progress? It has 900GB queued up and due to not having shares it's moving the data all wrong... Tools - diagnostics Do this before you reboot if you have to. All my shares are either cache-prefer or cache-no (so mover just starts and immediately stops because there's nothing to move), so not 100% sure if when mover is running you have the option to stop the array (which will effectively stop mover).
April 16, 20179 yr You can't stop the array when the mover is running. Can't find a way to get it to stop, tried "pkill mover" and "kill rsync". Tempted to just pull the plug and improper shutdown at this point. Edited April 16, 20179 yr by tyrindor
April 17, 20179 yr 1 minute ago, tyrindor said: You can't stop the array when the mover is running. Can't find a way to get it to stop, tried "pkill mover" and "kill rsync". Tempted to just pull the plug and improper shutdown at this point. There's only one file I see that mover tried to move (because it returned an error). And it was in the Movies share which is a use cache yes share (so it will move from the cache drive to the array), the allocation method for that share is most free, and split level is 0 and no disks are included / excluded on the share setting, so the file(s) will go to whatever disk has the most free space on it. Unfortunately with mover logging disabled (Settings - Schedule - Mover Settings), we can't see the log from mover for the files that you're saying its moving. OOM errors. Restart the system into safe mode. You've got some PhAzE plugins installed and something like nzbget can and will use all available memory to it when unpacking for instance. (That or you have multiple preclears happening -> preclear can also utilize a ton of memory with multiple clears consecutively
April 17, 20179 yr This probably explains what happened to User Shares, but I don't know what caused the oom. Apr 16 18:17:26 UNRAID kernel: Out of memory: Kill process 8106 (shfs) score 3 or sacrifice child Apr 16 18:17:26 UNRAID kernel: Killed process 8106 (shfs) total-vm:2114960kB, anon-rss:62456kB, file-rss:4kB, shmem-rss:1928kB Apr 16 18:17:26 UNRAID kernel: oom_reaper: reaped process 8106 (shfs), now anon-rss:0kB, file-rss:0kB, shmem-rss:0kB
April 17, 20179 yr Last time NZBGet got something was yesterday, no preclears going. I recently transferred 1TB of movies to my cache drive at 700MB/s over 10G about an hour ago. That's the only thing I could think of. Not many people have setups capable of those speeds. The mover is stuck, according to logs it's been sitting on a 10GB file for an hour now. Going to have to force shut down the server. There really needs to be a command to force stop the mover. Edited April 17, 20179 yr by tyrindor
Archived
This topic is now archived and is closed to further replies.