danioj

Thanks, managed to access the Web GUI in Bridge mode but couldn't in Host mode. When I access the WebGUI, under 'Server Network Settings' I see "eth0: 172.17.0.2" Should this match my internal IP range or external IP? as it does not match either. Also I have a bridge setup in UnRAID (br0) should I be using that instead of eth0? In short No, and default is for the Network Settings within OPenVPN-AS using this container to be (in this case this is my Backup Server who's IP address is 192.168.1.3): eth0: 192.168.1.3 As for an explanation, I shall do my best to explain as I understand it. No this is the "internal" IP of the Container. Note that you access this on the "unRAID Server" via its IP or hostname. E.g.: Container = 172.17.0.2:port# => unRAID machine = 192.168.1.1:port# AFAIK the port(s) in use by the container is usually static (in some cases changeable like OpenVPN-AS) but the port on the unRAID machine is variable. Note also that some Containers HAVE to run in Host or Bridge mode for them to work. Anyway, in general IF you select the Container to run in Host mode then you are essentially saying to Docker => please map the Container Port to my unRAID Server port. This can be a hassle IF you have more than one Container wanting the same port(s). This is where Bridge mode comes in. This is where docker differs from a VM for instance. It means that docker still uses the unRAID Server IP address, but that there is a mapping (bridge) between the ports as seen inside the docker container, and the ports these are mapped to at the unRAID Server level. This is how you can have two different docker containers think they are using the same port internally, but that they are seen as different ports on the unRAID Server. I hope that makes some sense to you.

I use the ID's as a whitelist. Currently, these are the current whitelisted id's: 4 5 9 183 184 187 188 190 196 197 198 199 The 9 (Power On Hours) and 190 (Airflow Temperature Celsius) are just for testing. Ah. Is there any reason we need to display 188 - Command Timeout for a preclear? I was under the impression that by-in-large it was advised that this should be ignored? I think even any change in this attribute is turned to ignore (as of 6.1.7 I think) by default in the WEB GUI too? http://lime-technology.com/forum/index.php?topic=44068.msg420647#msg420647

Looking good. I don't know about anyone else but when eyeballing a SMART report I find I use the SMART Attribute ID numbers. I guess they could be considered superfluous.

Using http://www.canyouseeme.org No! However, I CAN connect to the service! I AM connected to the service. Thanks, I'll keep at it. As a side note, if you are SURE that your setup is fine then you could try a different port. There is a possibility that your ISP (or something else) is blocking access to that port.

Using http://www.canyouseeme.org No! However, I CAN connect to the service! I AM connected to the service.

I can't see any obvious issues that would cause the test to fail. I have just tested both my main and redundant instances of this plugin though and both mine fail too. However, I have been using them all morning. I am even using one of them now. Have you tried to just connect anyway? The test says that: Perhaps the reason it fails has something to do with the fact it is running in a Container on a Server on a different Subnet with just ports forwarded to the connectivity ports. I don't know. I wouldn't worry - if it works - all is good! Give it a try!

Have you recently move the folder or done anything to manipulate the permissions? TBH it doesn't really matter. The last time I had this issue the ONLY way I found to restore the correct permissions was to delete the config folder. Delete the Container. Start again. I know that is probably NOT what you want to hear BUT it worked. Thankfully this Docker is REALLY easy to set up. And barring having to re-deploy my auto-login certificates again which took a little while, the actual setup only took 5 mins.

@gfjardim here is a nice thought regarding possible future features of the new BETA Plugin / Preclear Script from RobJ as a result of a support thread posted: http://lime-technology.com/forum/index.php?topic=47697.msg457468#msg457468 For your consideration.

Running at Full is raging loud, too loud for me. Is there an easy way to tweak the bios settings to different fan levels to see the speed change? Or do I need to set the level, save and reboot to see the change? Well, it depends what your board supports. For instance my IPMI interface only lets me select the following options: You should log into your IPMI interface and see what options there are to select. In this way you can "Tweak" without going into BIOS and rebooting etc. You could always NOT let the system get into unRAID (stop it at the boot screen) and just change the fan settings. Let it run for a while then look in the BIOS at temps. This is not a great suggestion BUT I am running out of them.

Saw your edit. Turn Smart Fan mode off. Set them to full. Or I guess you could play with the Smart mode temperature ranges BUT I would just get them running at full. The fans are quiet anyway.

Have to be quick as I'm on my way out (so excuse the poorly worded and brief response) but yes, get into your BIOS or IPMI interface (assuming it allows it) and set those fans to run at full speed! I know this function is coming in the IPMI Plugin BUT you don't need it. EDIT: however as I mentioned to another user earlier today, even in the 40's why worry? The new disks of today have acceptable operating temperatures up to ~70'C. Chill! (Pun intended)

Given most new drives have acceptable operating temps up to 70'c and you're getting temps of ~ 40'c under load. Well, I don't see that there is anything to be unhappy about.

Those temps under load?

All mine are 10" from an ASRock C2550D4I. Any longer and it would start to look messy! I agree, 8" would be fine for probably 3/4 of the ports (using the board I mention).

This ends up being more of a pain than its worth. Trust me, note down the servers and the important config and just start again with a new config folder for this container. It might sound and feel like a pain BUT you WILL end up spending less time in the long run!

That was quick!! ? Very easy fix. I don't have a system that I can set up two parity disks, so let me know if it worked. Like a charm. Let me know if you need anything else 6.2 related you need testing. No important data, pure test setup.

That was quick!! ?

I wanted to share that the current version of Unassigned Devices "seems" at first glance to be playing well with 6.2.0-beta18. Although interestingly at the point where the initial Parity Sync is happening the 2nd Parity drive still shows in the Unassigned Devices section even though it has been assigned to the 2nd Parity slot in the Array and the Array is started and a Parity Sync is underway: Yep, still there even when the Parity Sync has finished. Mount is greyed out though so at least you can't DO anything bad. UD probably needs to be taught to ignore that 2nd Parity I imagine. For giggles I clicked the preclear icon to see what happened: Didn't want to press Start

I think I was a little hasty in my apprehension about this work, so for that I am Sorry. After taking in all the facts (and also TBH some surprising behaviour by some of the longer serving members of the community) I would like to do my bit to support. I will use your BETA preclear version throughout my testing of unRAID 6.2 soon™ Edition through BETA -> RC -> FINAL: http://lime-technology.com/forum/index.php?topic=47427.0 I will work into what I am doing the ability to test as much as I can (e.g. between testing rebuilds etc) the following disks. And I have started them ALL off on a preclear run right now .... Parity 1: Western Digital Blue WD800AAJS (00PSA0) 80GB 7200 RPM 8MB Cache SATA 3.0Gb/s Parity 2: Western Digital Blue WD800AAJS (00PSA0) 80GB 7200 RPM 8MB Cache SATA 3.0Gb/s Data 1: Seagate Barracuda 7200.10 ST380815AS (9CY131-020) 80GB 7200 RPM 8MB Cache SATA 3.0Gb/s Data 2: Western Digital WD Blue WD800JD (60LSA5) 80GB 8MB Cache SATA 3.0Gb/s Data 3: Western Digital WD Blue WD800JD (60LSA5) 80GB 8MB Cache SATA 3.0Gb/s 6.2.0-beta18: *For those who notice, don't worry about the temperatures. As it is my test server with test disks I didn't double check the fans. They had failed. Replaced them as of the verify so temperatures are in order. Links below are screenshots from the Plugins' setup page and from Unassigned Devices on the Main Tab (if you're checking compatibility). UPDATE 1: Preclear on ALL drives was successful. Links below indicating status are once again from Unassigned Devices and Main Tab for results (including eye popup at finished) indication and indicating Start of an ALL disk Verify (including eye popup during). UPDATE 2: ALL Disk Verify on ALL drives was successful. Message reports back that the drive is now precleared. UPDATE 3: Then I tried to follow the instructions to use the exisiting preclear plugin to run the Test and Postread Verify from the older plugin. Unfortunately I could not. After placing the Joe.L version (original) version of the script on the flash, here is what happened: Firstly I note that once I pressed the X it cleared the status BUT now no-where does it indicate that the drive is preceared. We should probably preserve that pre cleared flag and indicate it somewhere on the GUI. Then I tried to run the Test and I got a blank status field in the GUI (which was not helpful). I had to click the eye to see what had happened. Of course, the stupid disk.cfg issue. Can we resolve this? After adding a disk to a slot, starting the Array, stopping the Array and then removing the disk (which meant I now have a disk.cfg) I tried to run a Test again. I was once again I got a blank status field in the GUI (which was not helpful). I had to click the eye to see what had happened. This was the -R issue reported by johnnies.black and BRiT indicating that the version of "sfdisk" included in unraid 6.2.0 beta 18 is different, and doesn't seem to support the uppercase R option. Not knowing where to go from here I decided to try the next step, which was the Run Post Read Verify. Unfortunately I was not able to complete this either. The popup window opened, it allowed me to select "Run post read verify" but when I clicked Start nothing happened. It just sat there. At which point the GUI also hung for a while. At this point. Mainly due to the HANG I went to the CLI and looked at the syslog. It appeared unRAID was spamming my Syslog with messages about the disks. I wasn't sure if what was happening was due to the preclear or something going on in unRAID. I made an educated (although it might have been a poor one) decision to decide it was something happening in unRAID (don't yet know if it is harmless) so I grabbed the diagnostics file and posted about it here: https://lime-technology.com/forum/index.php?topic=47408.msg453990#msg453990 With only assigning the disks to the Array left and doing a Check Parity I decided to take stock at what the failed attempts at the above operations had done to the status in the Preclear GUI and the Unassigned Devices GUI and this was the result. Messages and Status' were confused. It also didn't seem to grasp that the Seagate had been precleared either. UPDATE 4: I started the Array fine. Formatted the disks (nice to see the default is XFS now as I didn't check this) and started a Parity Sync. UPDATE 5: Finished (without step 4 as indicated above). Parity Sync Complete. unRAID 6.2-beta18 seems to have been fine with the work that the Plugin has done. Dual Parity Protected Array** **Note: For those that are 6.2 conscious, I have noted how the 2nd Parity Disk appears in the Unassigned Devices area and have posted this in the Unassigned Devices support thread here: https://lime-technology.com/forum/index.php?topic=45807.msg453974#msg453974 gfjardim: Please let me know if you would like anything else tested / done. Hope this helps.

He won't give you the authorisation or you can't get hold of him!? Haven't you incorporated both scripts (Original and Faster) into the Plugin already? Have you been asked to remove them? If it is a matter of contact I am SURE someone can help facilitate that. If it is the latter then I would be SHOCKED. Your Plugin has added much needed front end to an almost ESSENTIAL task. I won't speculate further until you confirm. Is there perhaps another reason WHY you are re-writing it too? Are you looking to add further functionality? If so what is this functionality? If at all possible (if you really don't have to) I think given the exposure your plugin gets via Community Applications - not to mention how well you have done at making a CLI exercise task so easy and user friendly via the GUI - I think it should only use one of the two very well tested scripts. Although it is officially unsupported by LT, a preclear is a task that almost every user of unRAID does. All new users of unRAID are advised to do it by Staff, Mod's and Experienced Users alike. At the moment the ability to do it is STABLE. Your Plugin is great as all it does is add a front end to those stable scripts whose (once executed) outcomes are very assured. My issue is that by introducing an untested scripts into the mix does a few things: - Removes confidence in the preclear script; - Introduces possibilities for error / bugs etc; - Add's confusion etc Now, like I said above, IF you are being told you can't use the scripts then I guess you have no choice BUT I think that would be a big step back from where we are today. If it isn't broken don't fix it. Can you add some more context to your decision?

Does anyone know where "fuse.conf" is placed AND if it is persistent? I am trying to edit it to enable edit fuse.conf to include user_allow_other so I can pass the option allow_other shfs option but AFAIK it will be ignored until you also edit /etc/fuse.conf to include user_allow_other.

Perhaps. I haven't used screen since the earlier version of the Plugin. I updated the plugin without taking head of release notes clearly AND didn't know I had to go in and switch them on. All good now though. Thanks for the explanation.

Have you actually gone to Settings->Nerd Pack and checked that screen has been selected for install? Earlier versions of the plugin installed all the options, but a while ago it was upgraded to only install the selected options. Very strange. When I went to Settings > Nerd Pack, all the tools were Off. I turned them on... and hit Apply. Not all turned on. At least screen which is the one I most need is On. Thank you guys for pointing me in the right direction. H. I just had this issue too. Exactly the same. I went to run a script in a screen session and the command wasn't found. Checked the Plugin Status, Installed. Checked my Backup Server and the same thing happened. I start to think it must be the Plugin given screen wasn't available on 2 different servers. Only when I saw your post did I think to check on the settings page... Same as you ALL of the packages were set to OFF on both the Main and Backup Systems. I have enabled them now and, like you, I now have screen running. Not sure how long my server has been running with these off or what caused the issue. However given it happened on 2 seperate servers I can only guess the issue is systemic. Interestingly there are also 2 packages which won't enable: - cksfv-1.3.14-x86_64-1_slack.txz - fdupes-1.51-x86_64-3_slack.txz

None at all or just reporting that no bytes trimmed?

IMHO the Noctua NF-F12 PWM fans are Perfect. I have them in my Setup and they are awesome. http://noctua.at/en/nf-f12-pwm http://www.newegg.com/Product/Product.aspx?Item=N82E16835608026