September 8, 20169 yr I think, for example, that in the case mentioned r.e. the turbo-write setting, the OS should base its choice on whether or not all disks are currently spinning, which to me is the intuitive meaning of "auto" in that case -- but that's also not the case [in that instance, auto = Off]. That's LT's plan, to enhance that in the future, and by having it default to auto now, when users update they will automatically take advantage of that feature.
September 8, 20169 yr I think, for example, that in the case mentioned r.e. the turbo-write setting, the OS should base its choice on whether or not all disks are currently spinning, which to me is the intuitive meaning of "auto" in that case -- but that's also not the case [in that instance, auto = Off]. That's LT's plan, to enhance that in the future, and by having it default to auto now, when users update they will automatically take advantage of that feature. This whole issue is coming about because all of you are assuming that "Auto/Automatic" is a noun. It is NOT! It is a adjective! It must modify something. When Auotmatic becomes subverted it to a noun, it can mean whatever anyone says it means until by general consensual (which will happen at some point) its meaning is firmly establish in the collective.
September 8, 20169 yr Why mark it "experimental"? Mainly because I didn't want to take a bunch of grief for adding a "new feature" in an RC release. But also, normally with changes like this we like to run on a wide variety of h/w configs, just to "make sure" there are no hidden gotchas. In this case, this is an option which improves performance for a certain customer config we were testing. In previous testing we used the 'boot/extra.cfg' file to add this option, but in the interest of making it easier on a user we decided to just make it an official config setting. Seems to me it isn't a "new feature" introduced in the RC, but rather a Change in how it is accessed / activated. It should have been listed in the change log at the point that it was first possible to activate it through 'boot/extra.cfg'.
September 8, 20169 yr Upgrade from rc4 went without any issues. One thing I noticed though: When I try to login on the web GUI with any other user than root it simply won't work. I also tried logging in via SSH. If I try to login with any other user than root I get the following console output: ----------------- Ex0danify@SKYFIRE:~$ ssh [email protected] [email protected]'s password: Linux 4.4.19-unRAID. Connection to 192.168.2.100 closed. ----------------- Anybody else with this problem?
September 8, 20169 yr Upgrade from rc4 went without any issues. One thing I noticed though: When I try to login on the web GUI with any other user than root it simply won't work. I also tried logging in via SSH. If I try to login with any other user than root I get the following console output: ----------------- Ex0danify@SKYFIRE:~$ ssh [email protected] [email protected]'s password: Linux 4.4.19-unRAID. Connection to 192.168.2.100 closed. ----------------- Anybody else with this problem? Fairly sure unRaid has always operated like that. Only root can access the webUI. And since at least the v5? days root has no access to smb user shares.
September 8, 20169 yr I hear from the community that there are a number of changes to KVM in the latest distro's? I assume these will make it in to the 6.3.x betas? I have no issues with 6.2 atm but as it stands, KVM in UNRAID isn't a patch on vmware's resource management.
September 8, 20169 yr Author I hear from the community that there are a number of changes to KVM in the latest distro's? I assume these will make it in to the 6.3.x betas? I have no issues with 6.2 atm but as it stands, KVM in UNRAID isn't a patch on vmware's resource management. Yes that's the plan.
September 8, 20169 yr Any suggestions on what I could do to try and resolve my problem? File copy speed/performance is fine under RC3, upgrade to RC5 and it slows down to a crawl.
September 8, 20169 yr Any suggestions on what I could do to try and resolve my problem? File copy speed/performance is fine under RC3, upgrade to RC5 and it slows down to a crawl. More details will be needed. Like, which way (to-or-from, or both), the actual change in speeds, OS's involved, all computers or just one. Oh, and that requested diagnostics file (read first post in thread for details).
September 8, 20169 yr The details are on page 5 of this thread along with a diagnostics file from when I was on RC5, once I switched back to RC3, the copy speeds returned to normal. My server has both an onboard intel nic as well as a card, currently I am only using the card, not the integrated. The integrated model is:Intel Corporation 82579V The card is:Intel Corporation 82574L My mother board is:ASUSTeK Computer INC. - P8Z68-V PRO
September 8, 20169 yr The card is:Intel Corporation 82574L My server uses the Intel Corporation 82574L card, without issue as far as I know. I believe many others use it too. I don't think the Intel is the cause. Have you tried tweaking it with the Tips and Tweaks wiki page (use the Tips and Tweaks plugin)?
September 8, 20169 yr No I haven't tried the plugin, I just installed it so maybe at some point I'll upgrade to RC5 again and see if any of the tweaks work. Just seems odd to me that since running the betas and subsequent RC's that all of a sudden I have this network performance issue with RC5
September 8, 20169 yr @ashman, have you tried disabling Direct I/O in network settings. Currently its set to Auto, try changing to 'off'.
September 8, 20169 yr @ashman, have you tried disabling Direct I/O in network settings. Currently its set to Auto, try changing to 'off'. Auto=off
September 8, 20169 yr @ashman, have you tried disabling Direct I/O in network settings. Currently its set to Auto, try changing to 'off'. Auto=off Unless the OS decides to turn it on... worth a shot since RC3 works and RC5 is causing him problems... when was that Direct I/O feature added again ?
September 8, 20169 yr Problem! I updated from 6.1.9 to 6.2 RC5, let it update for a while, but nothing happend. I could ping the server, and ssh into it, but no webui response. Tried shutting it down from a ssh session - Nothing happend, tried pressing the power button case, same thing - nothing happend. I typed mc in the command line, just to see, but nothing was mounted at /mnt I could also log in into the server without any password with the root user. I forced a shutdown of the server after a long time, since I had no other choice. When it boots now, it spams "Waiting for /dev/disk/by-label/UNRAID (will check for 30 seconds)" for 30 seconds, then it does not find it and continue booting. My flash drive is still named UNRAID, and I have also tried running "make_bootable.bat" again just to see, but same thing happens. Todays part of the syslog attached, and a diagnostic from today (probably from powerdown plugin?) The standalone syslog shows more info than the syslog in the diagnostic file Any help would be appreciated! syslog.txt tower-diagnostics-20160908-2144.zip
September 8, 20169 yr Not 100% sure if this issue resides with 6.2RC5 but I have never had this problem before. I moved from 6.1.9 to 6.2 at RC3 and still no problems until now. I have been trying to get a new docker setup this evening and twice my array disappeared and I got "Transport endpoint is not connected" errors. The only solution I could find was to restart my server. Diagnostics attached: Many thanks for the continued work! tower-diagnostics-20160908-2209.zip
September 8, 20169 yr It has now happened a third time this evening. As stated earlier, this has never happened before and the only solution I have is to restart my server. Every time it happens I am interacting with a Docker container, I am trying to fix. (during this process the container has been complete removed and re-added). Attached is another set of diagnostics. tower-diagnostics-20160908-2305.zip
September 8, 20169 yr It has now happened a third time this evening. As stated earlier, this has never happened before and the only solution I have is to restart my server. Every time it happens I am interacting with a Docker container, I am trying to fix. (during this process the container has been complete removed and re-added). Attached is another set of diagnostics. Is it a particular container that's triggering it or any at random?
September 8, 20169 yr not really 1 single container but 1 application: I have been having problems with linuxserver:deluge since its move to alpine so I was trialing needo:deluge and also binhex:deluge to see which would be better while also offering newest build of deluge 1.3.13. The problem happened with both binhex:deluge (x2) and linuxserver:deluge (x1) after returning to linuxserver:deluge to see if binhex:deluge could possibly contribute to the problem. It has happened 3 times in total. I then removed my deluge container outright for the night unusual I know
September 9, 20169 yr Drop down out of the network setting for the subnet mask does not give you the option for a 255.0.0.0 subnet, it only has 255.255.0.0 Do you need a /8 network (16.7M addresses) ? Yes, I know my network sub-netting...unless this is a limitation in the programming.
September 9, 20169 yr Problem! I updated from 6.1.9 to 6.2 RC5, let it update for a while, but nothing happend. I could ping the server, and ssh into it, but no webui response. Tried shutting it down from a ssh session - Nothing happend, tried pressing the power button case, same thing - nothing happend. I typed mc in the command line, just to see, but nothing was mounted at /mnt I could also log in into the server without any password with the root user. I forced a shutdown of the server after a long time, since I had no other choice. When it boots now, it spams "Waiting for /dev/disk/by-label/UNRAID (will check for 30 seconds)" for 30 seconds, then it does not find it and continue booting. My flash drive is still named UNRAID, and I have also tried running "make_bootable.bat" again just to see, but same thing happens. Todays part of the syslog attached, and a diagnostic from today (probably from powerdown plugin?) The standalone syslog shows more info than the syslog in the diagnostic file Any help would be appreciated! Issue fixed. Would be nice if LT could take a look at this to prevent it in the future if possible. https://lime-technology.com/forum/index.php?topic=51715.0
September 10, 20169 yr - add experimental global share setting: "Tunable (enable Direct IO)" Update on this setting, I've been using it since rc5 was release without any issues, array was up for 5 days, but some time a go (maybe 1 or 2 days) speed reverted back to what I used to get with the setting off, <800MB/s, stopping and starting the array fixed and speed is >1GB/s again, will test every day now to check if it happens again.
September 10, 20169 yr Two issue which change in RC5 1. If any disk missing in array (no parity), it not allow start because "too many error", in previous version no this limitation. As one of my unRAID server use for backup purpose and I always insert different disk(s) set. Now I need doing "new config" and re-assign all disk, this make trouble. 2. When browsing "share" or "disk" in console, it won't indicate the directory/file in which disk now, but previous version will show.
September 10, 20169 yr 2. When browsing "share" or "disk" in console, it won't indicate the directory/file in which disk now, but previous version will show. i am having no trouble doing this. Perhaps I have misunderstood what you mean, so you could provide a screenshot to illustrate your issue.
Archived
This topic is now archived and is closed to further replies.