bonienl
-
Posts
10233 -
Joined
-
Last visited
-
Days Won
65
Content Type
Profiles
Forums
Downloads
Store
Gallery
Bug Reports
Documentation
Landing
Posts posted by bonienl
-
-
By the way, bonienl, I believe you indicated that a way to call for the local generation of the diagnostics dump from the command line was coming. If it's ready now, can you enlighten us how to do it? I'd like to add that to the 'Need Help' post, for when the network is down.
The script diagnostics is used to collect the information. The webGUI calls this script in the background with one parameter: the name of the zip file.
When telnetting into the system (e.g. when the webGUI is non-responsive), the script can be executed without any parameters. This will produce a zip file with the same naming convention as used in the GUI and stores the file under /boot/logs (the folder will be created if not existing)
*Note for LT: still need to symlink the diagnostics script so users can simply type the script name without the full path.
And while I have you, a few more requests? Could you change the 'array' folder name to 'system', keep array.txt in it, add system.txt with the results of 'lspci', 'lsscsi', and 'free -mt' (maybe others later?), and add network.txt with the results of 'ethtool -i eth0', 'ethtool eth0', and 'ifconfig'. Of course, you may have better ideas!
I'll have a look at that, thanks for the suggestions.
-
How to reproduce:
Change rapidly between WebGUI tabs back and forth as if you were (e.g.) searching something.
I did very rapid page retrieval and can NOT reproduce this problem in rc4. emhttp stays solid while getting requests fired at it.
Do you have plugins, dockers or VMs running as well ?
Did you scan your syslog for 'oom' (out-of-memory) messages ?
-
I've noticed on RC4 the page update frequency feature set to real time does not dynamically update my drives read/writes until i refresh the webpage. This is from multiple pcs/browsers. Any ideas?
Read/Write counters are updated every minute, this hasn't changed.
I have performed a test by transferring 250gb to disk3 in my array and waited 15 minutes with disk 3 still showing as spun down. After a manual refresh on my chrome tab I then see disk 3 spun up. Of note before upgrading to RC4 I formatted my USB key, installed Unraid and copied my config folder over. Is there something legacy within my config folder possibly causing the issue?
I tested this and it is working as expected in RC4. Counters and disk status are updated every minute.
Check if you have anything installed in /boot/extra, /boot/plugins, /boot/config/plugins.
-
I can confirm that apcupsd is starting up correctly in rc4, and that notifications are working again.
Thanks for testing and confirming.
-
This happened to me on a previous upgrade. I ended up finding an old log window (which had worked) was still open. After closing the old log window, a new log window worked fine. I tried to reproduce the problem but couldn't.
Yeah indeed opening more than one popup window will result in empty windows, only the first window receives information.
-
Have you tried to clear your browser cache (shift-ctrl-del) ?
Yes, and same result.
I'll try to investigate further as this is kind of annoying
My logs are working ... this happens to all your installed Dockers ?
-
flash drive share settings should also be moved to the now unified (almost) shares page. IMHO
No flash is a special case. Export function & settings need to be available all the time.
User shares and disk shares on the other hand are not available when the array is stopped.
-
Has anyone else lost the ability to see Docker logs? It opens the popup window, but never receives any data. Tried running the command-line version, and didn't get any output either (I've left the popup window up for an hour+ and no data filled).
I'm experiencing the same issue here since the upgrade from RC3 to RC4.
1. Dashboard view, opening the logs of a VM or Docker opens an empty window that never gets filled
2. Dashboard view, click on unraid log opens an empty window as well
3. Docker and VMs views, opening logs has the same result, a white window with no info
Command line does work
/usr/bin/tail -n 42 -f /var/log/syslog 2>&1
/usr/bin/docker logs --tail=350 -f PlexMediaServer 2>&1
...
Tested with Firefox 38 and IE 11
Have you tried to clear your browser cache (shift-ctrl-del) ?
-
I've noticed on RC4 the page update frequency feature set to real time does not dynamically update my drives read/writes until i refresh the webpage. This is from multiple pcs/browsers. Any ideas?
Read/Write counters are updated every minute, this hasn't changed.
-
It looks like you are viewing your unRAID GUI through unMENU? If yes, maybe test the view through unRAID direct?
Mine doesn't do that and I am using Chrome on Win 8
Good pickup archedraft, I just assumed he'd changed the port..
Yes this is happening because the GUI is viewed from within unMenu. Use native display instead.
-
the diagnostics tool still includes my email address in the syslog file.
Jun 3 07:29:32 Unraid-Nas emhttp: /usr/bin/tail -n 42 -f /var/log/syslog 2>&1 Jun 3 07:38:22 Unraid-Nas php: /boot/config/plugins/dynamix/notifications/agents/Pushover.sh Jun 3 07:38:23 Unraid-Nas php: {"status":1,"request":"xxxxxxxxxxxxxxxxxxxx"} Jun 3 07:38:48 Unraid-Nas sSMTP[26761]: Creating SSL connection to host Jun 3 07:38:49 Unraid-Nas sSMTP[26761]: SSL connection using xxxxxxxxxxxxxxxx Jun 3 07:38:51 Unraid-Nas sSMTP[26761]: Sent mail for xxxxxxxxxxxxxxxx (221 2.0.0 closing connection fs9sm30173301wjc.34 - gsmtp) uid=0 username=root outbytes=641 Jun 3 07:42:36 Unraid-Nas kernel: docker0: port 2(veth67c8319) entered disabled stateobviously i replaced the actual email address with xxxxxxxxxxxxxxxx
and also the pushover request and some sha256 reference (wasn't sure what they were, but i xxxxx'ed them anyways)
Good points.
I look into sanitizing the syslog before saving.
I'm not crazy about sanitizing the syslogs, but will accept it if that's what others want. First, it slows the process down, second - it does nothing for those other methods of capturing the syslog, and third - why is it that important, what abuse is possible? And would it not be better to sanitize at the source, when necessary? As in make sure that email and notification tools are careful what they log.
I agree it's good as a general policy to limit email address exposure, but I'm not convinced there's a vulnerability here yet, when they are only exposed in attached files, not in text on a web site. Happy to learn from others...
For my part, I'm completely committed to protecting all users privacy, will do all in my power to never take advantage of anything I may inadvertently notice. And I trust and hope all others will do the same. Is that enough, I don't know.
Off-topic but related, I have never seen the point in hiding drive serial numbers. While there are serial numbers that are associated with private stuff, many serial numbers are just unique identifiers, that's all, and useless for anything else. Hard drive serial numbers fall into that last category, and I'm completely happy to expose mine. But I certainly don't know everything, so it's *almost* conceivable that a hacker somewhere has figured out how to take advantage of someone else's drive serial number. Can't imagine for what though, so until I do, I won't hide them.
I have added the following code to the diagnostics script:
sed -r 's/\S+@\S+\.\S+/[email protected]/;s/(username|password)([=:])\S+/\\1\\2xxx/' $file|todos >/$diag/log/".basename($file).".txt"
This would change an email address into [email protected], and username=xxx, password=xxx (if existing)
Doesn't really slowdown the file generation ...
-
I have a GUI question...
Should the information on pages not try and re-locate based on window size? For example, if I am on the Main tab, and Array Operations and I choose to either Stop the Array or Spin Up/Down disks I have to scroll down through a bunch of whitespace before I see the actual activities at the bottom of the window. Should this not treat the bottom of my browser window as the bottom of the page and display the info there? It seems like the page has a defined length that is independent of the amount of data on the page, or the size of the browser window.
It's minor, but it's somewhat annoying.
Hopefully this makes sense.
All information is placed immediately at the top without additional whitespace. What browser are you using and is it PC or smart-phone/tablet ?
-
Has anyone else lost the ability to see Docker logs? It opens the popup window, but never receives any data. Tried running the command-line version, and didn't get any output either (I've left the popup window up for an hour+ and no data filled).
Docker logs work for me as before, i.e. info present.
-
the diagnostics tool still includes my email address in the syslog file.
Jun 3 07:29:32 Unraid-Nas emhttp: /usr/bin/tail -n 42 -f /var/log/syslog 2>&1 Jun 3 07:38:22 Unraid-Nas php: /boot/config/plugins/dynamix/notifications/agents/Pushover.sh Jun 3 07:38:23 Unraid-Nas php: {"status":1,"request":"xxxxxxxxxxxxxxxxxxxx"} Jun 3 07:38:48 Unraid-Nas sSMTP[26761]: Creating SSL connection to host Jun 3 07:38:49 Unraid-Nas sSMTP[26761]: SSL connection using xxxxxxxxxxxxxxxx Jun 3 07:38:51 Unraid-Nas sSMTP[26761]: Sent mail for xxxxxxxxxxxxxxxx (221 2.0.0 closing connection fs9sm30173301wjc.34 - gsmtp) uid=0 username=root outbytes=641 Jun 3 07:42:36 Unraid-Nas kernel: docker0: port 2(veth67c8319) entered disabled stateobviously i replaced the actual email address with xxxxxxxxxxxxxxxx
and also the pushover request and some sha256 reference (wasn't sure what they were, but i xxxxx'ed them anyways)
Good points.
I look into sanitizing the syslog before saving.
-
Haswell CPU speed reporting seems to be much improved. Happily showing correct speed it seems.
Some improvements were made in that department, thanks for confirming !
-
So, what about the diagnostic utility plugin we were testing? Should we uninstall it now?
Yes, the plugin is superseded and should be uninstalled (you will need to reboot to get the built-in function though) - sorry it wasn't mentioned.
-
In your examle you use twa0. How is the correlation between device "sdb" and "twaN" ?
See we have device "sdf" what will "twaN" be or do you use some lookup table ?
-
Yes I think I noticed this on a previous release so I don't think it is new with RC4The More button is added when you have the optional Dynamix System Information plugin installed, clicking on it brings you to that tool.
Just a little nuance ... previously the Close button was replaced by More, now there are two buttons.
-
The More button is added when you have the optional Dynamix System Information plugin installed, clicking on it brings you to that tool.
-
Apparently that is a new Tool under 'Tools' called 'Diagnostics'. It will collect all manner information on your system including syslog, smart reports, Shares information, array information and most of the configurations files zipped up into one containing file. It should be a really great help in getting the information necessary to help people with issues.
(I got it a few weeks as part of a testing group and I saw today where someone had included it in a post. I queried about it and he said he just found it and thought it "Great improvement of RC4 ".)
Good to hear, this is useful
-
I thought this very minor reporting issue was mentioned in earlier betas but I still see it in RC4.
Max/Installed memory seems buggered...
Also, should the system information report on the total # of physical CPUs (I have two)?
John
Can you telnet into your system and post the outcome of these commands.
dmidecode -q -t memory
dmidecode -q -t 4
Thanks
-
(3) The Help button is nice; but it'd be nice if it "reset" to a cleared state when you switched tabs. I was on the Dashboard and clicked Help to see what it would do ... answer is nothing. Then clicked it again and it still did nothing. Then clicked it a 3rd time and finally decided it really didn't do anything. Clicked to a different page and Voila!! There was help text displayed. A few seconds of experimenting and it's obvious that the button simply turns it on/off and if it's on then you'll see the help on any tab that has help. I think it would be more intuitive if, when you were on a tab and needed help, you simply clicked Help. You could then turn it off (as now) by clicking it again ... or you could just go to another tab and it would automatically be turned off.
I don't mind the way it works right now, but the help button would make more sense as a toggle like the "Advanced/Basic" view toggle on the docker tab.
It is a toggle the way it's working. What's not intuitive about it is that if you toggle it on while viewing a tab that has no help, nothing changes (except the "Help" button is visually slightly depressed). It's clearly not a big deal ... but I think it'd be more intuitive if it reset whenever you changed tabs. I suppose a "Default Help State" setting could control whether that "reset" was to an on or off state
Personally I find it very annoying to switch on every time 'help' when a page is changed, when I want help then I want it to be there all the time, again my personal view.
When every page has some help information it would be clearer that help mode is on or off (besides the highlighting of the button), could be an objective
-
+1
Looking for an interface similar to what we get with Dynamix to manage Docker containers
Have a look here http://lime-technology.com/forum/index.php?topic=40262.0
-
Well, official version 6 isn't that far away. If you have a little patience you will be able to upgrade to v6.
[Solved] unRAID 6.0 RC3-4 Dynamix WebGUI OFTEN hangs (solved in RC5?)
in Defect Reports [DEPRECATED]
Posted
With first two tabs you mean "Dashboard" and "Main" ?
When this happens can you list all running processes: ps -ef
And communication state: lsof -Pni