"SimpleFeatures" Plugin - Version 1.0.11

[Sccrluk9]

Request for the next version. Could we get it so the cache drive is not included in our totals? Maybe put it below the totals like the flash drive is?

 

Edit: After looking a little closer it seems that only read and write counts are added to the total. I still think separating it out would make more sense but its a little less of a concern now.

[Sensei73]

someone here also having problems with "Network idle threshold" in sleep mode? I cannot same any value.

[HarryRosen]

anyone having slow response times? or the page loading really slow, just installed 1.011 and I can barely load a page. I ended up moving back to 1.0.5 and that works fine. Any ideas

[peter_sm]

One question about SF S3 (sleep)

 

When I have my OpenVPN tunnel active (tap0) the s3 script dosent let my sever sleep, I have tested this outside SF and run without eth0, then everything works both when openvp is enabled and disabled, so can this be change so it's possible to disable etho?

 

Thanks

 

Peter

[progrocktv]

I've been trying the sleep command from the SF 1.0.11 with no luck. Upon clicking SLEEP the system locks up and can no longer access the Web GUI. After a manual restart I received the UNCLEAN SHUTDOWN notification & the system goes into a parity check once the array goes back online. I've checked and re-checked the BIOS setup & recently installed an Intel Pro 1000 and could get it to sleep, but no wakeup. Now suddenly it just hangs on sleep.  Any help would be appreciated (syslog attached & hardware noted in my sig)

 

Thanks!

syslog.txt

[EdgarWallace]

First of all: thanks to the developers for the great addon to unRAID.

 

I installed SF from the very beginning and it's getting better and better and even easier to install. The question that I have it related to the preclear feature. I even can't find the file at the google page, hence the error is pretty obvious, or am I wrong?

 

installplg /boot/plugins/simpleFeatures.disk.preclear-1.0.10-noarch-1.plg
installing plugin: simpleFeatures.disk.preclear-1.0.10-noarch-1
file /tmp/plugin-prepare: successfully wrote INLINE file contents
  /bin/bash /tmp/plugin-prepare ... success
file /boot/plugins/simpleFeatures/simpleFeatures.disk.preclear-1.0.10-i486-1.txz: downloading from http://unraid-simplefeatures.googlecode.com/files/simpleFeatures.disk.preclear-1.0.10-i486-1.txz ... http://unraid-simplefeatures.googlecode.com/files/simpleFeatures.disk.preclear-1.0.10-i486-1.txz:
2013-03-03 13:08:54 ERROR 404: Not Found.
bad download, deleting

[mikejp]

 

I installed SF from the very beginning and it's getting better and better and even easier to install. The question that I have it related to the preclear feature. I even can't find the file at the google page, hence the error is pretty obvious, or am I wrong?

This thread explains it ...

 

http://lime-technology.com/forum/index.php?topic=26121.0

 

[EdgarWallace]

Thanks a lot. That is really a pity. Anyhow, the plg should then be removed from the ZIP file.

 

Another question that I have: what is the latest mySQL version that is running without any hassle? I still have installed Mysql-5.1.66-i468-6atv.plg.

 

Is anyone here happy with an more recent update?

[CHBMB]

Thanks a lot. That is really a pity. Anyhow, the plg should then be removed from the ZIP file.

 

Another question that I have: what is the latest mySQL version that is running without any hassle? I still have installed Mysql-5.1.66-i468-6atv.plg.

 

Is anyone here happy with an more recent update?

 

Here you go. Unfortunately the server where the package is located updates so the plugin is only valid for the version it was created for and that version is removed once there's an update, so recommend saving a backup of the package file that is downloaded.

 

http://lime-technology.com/forum/index.php?topic=20603.msg229248#msg229248

[mobias1313]

I have been having an issue where trying to do a reboot or a clean shutdown will completely shut down without me logging in to the server and typing power down. It seems to get caught up on shutting down mysql. Has anyone else had this issue?

[mark_anderson_us]

Hi Guys

 

I really like this UI, but had huge problems connecting on port 80. 9 times out of 10, I got could not connect in browser. Am I supposed to disable the built-in WebUI or move it to another port? If so, where is that done? Could I configure the standard WebUI/SimpleFeatures to use port 81? If so, how?

 

unMenu still worked great on port 8080.

 

Coincidentally my ssmtp stopped working. I've disable all the SimpleFeatures packages for now and the standard WebUI and ssmtp now work again

I saw segmentation faults in syslog related to emhttp

 

Finally, in the poll at the top of the post, there is an option that says "Can only be implemented once V5 removes emhttp". Can anyone elaborate on this? I thought emhttp was the web server for the original WebUI. (new to unRAID, so could have it totally wrong).

 

I'd like to reinstall simple features, but want access to the original (supported) WebUI. I cannot live with the high rate of the WebUI's being unreachable.

 

Regards

 

Mark

 

[Helmonder]

Simplefeatures is more like a different "skin" (with added functionality) running on top (and replacing) the unraid stock interface. That is a different setup then unMenu where it runs in a total seperate process.. So what you want will not be possible unless there will be some large scale changes to the basis of SimpleFeatures, and those are not expected I think.

[kricker]

Sorry, but  I still cannot get the sleep script to work. Any ideas for how to troubleshoot it? It doesn't even work when I start it from the commandline myself with the parameters -a -c -m60. Any advice?

I'm not having any luck with the sleep feature either. I'll probably have to do it manually in the GO file again. Bonienl , speeding_ant, anything we can do to help troubleshoot this?

 

Having another look it appears something is keeping one of my disks spinning. Maybe if I can figure that out it will work.

[Joe L.]

 

The plugin released has modified that script, and although it still seems to work fine, it is still changed, and is not the product that Joe originally created and volunteered to support.  ...............

And it contained a version with an identical version number to the one in my development system.    Yes, I could leap-frog it, and go to the next, but that is not a game I wish to play.

It's not that Joe is peeved that his work was 'stolen', more that he's on the hook to support changes he knew nothing about, have circumvented his safety measures and that he wasn't even made aware that it was going to happen.

 

Sorry, I strongly disagree with this sort of thinking and I am thankful most people also see the nonsense in thinking this way, including Lime. If Lime thought the way you do, then we would have no plugin interface at all.

The entire ability of unRAID to be extended with .plg plugins is because I and others collaborating with unMENU plugins proved there was a desire to use the server for more and to improve on its interfaces.  Yes, unMENU has had plugins since the very early days.  Many were contributed by others, and we worked together in suggesting improvements and additions.  There was only one user-contributed plugun I did not include, but only because it had a sub-set of the functionality I had in a plugin I wrote but had not yet released.  (It was released instead)

I highly doubt the simple features plugin directs people to the original support thread (but I admit I have not used it to verify). I'd like to see where you got the idea Joe would be required to provide support to people who did not use his original preclear script.

The SF modified version of my script, but with the same/overlapping version number, and the same name, support would be a nightmare.  It is for that reason I objected.  I did not agree with the removal (bypass) of one of the safeguards to protect a user's data.  That I feel must not be removed.

 

First off there is no plugin interface the plugin manager that was promised never came about (as it stand right now).

True, and a user poll asked that Tom concentrate on getting 5.0 released and stable prior to his creating a plugin manager.

See this thread: http://lime-technology.com/forum/index.php?topic=19790.msg175898#msg175898

Guess where all the questions, complains all went? Check JoeL's support thread, they all went there.

And as upset as he was, he still helped everyone. So go and try to taint him some more...

Thank you for understanding my view point. (even if others do not)

I understand it's pretty easy for you to want what YOU want, and have others deal with what becomes of it. After all you must have donated to JoeL and SF

I've only twice had a PM asking where a person could donate towards my efforts.  (and can't remember any in the past year or so)  I know I've had a few others offer a drink or a meal if I'm ever in their area, but that is about it.   

 

It is not the money.  I'm not independently wealthy, but not terribly in debt either, even though I'm not currently working.  (In both cases, I politely asked the person to pick a local charity, or church, and donate the money to them in my honor, even if it was only a few dollars, someone needy would benefit. I figure a poor person needs the donation more than me.)

and written some scripts and plugins of your own or at the very least pitch in. Wait, what's that u say?

Most do not have the experience to pitch in with code of their own.  Many pitch in with suggestions for improvement.  It is big mistake to ignore them.  Some may not be possible, but others are, and that is what collaboration is all about, and what was missing in this case.

 

as mentioned above, my "cache_dirs" script was also included in a different plugin, again without any coordination or request for an OK, or contact of any kind by the same author... It is unchanged (I think) and I was not really happy about it overwriting a version I might subsequently distribute, I kept quiet.    The support request did come to me, and one of the things I had to do first is figure out if it had been modified in any way by being included in a plugin that would have caused the bug being reported. 

 

I have been contacted via PM by the original author of SimpleFeatues about this, but not from bonienl.  bonienl did recognize his error and apologized here: http://lime-technology.com/forum/index.php?topic=26121.msg228195#msg228195

 

I've little/no free time to work on my improved version of the preclear script at this point.  In a few weeks perhaps...

 

In any case, thanks for the kind words...

 

Joe L.

[speeding_ant]

I can see that simpleFeatures is starting to reach its limits with the current built in web server. Emhttp wasn't designed with multiple consecutive http requests in mind. We even have to parse GET ourselves (POST won't work) to get anything to work the way it does.

 

I've yet to see conclusive evidence that simpleFeatures is causing crashes (on our setups we've never had a single issue, with up times reaching months). The majority of users seeing these problems are also running other plugins such as SABnzbd.

 

It is rumored that Tom will be including lighttpd in unRAID 5.1, to help incorporate a plugin manager - and to help out plugin developers such as ourselves. At that point we can make a conscious effort to remove legacy code to provide a rock solid environment.

[Joe L.]

I can see that simpleFeatures is starting to reach its limits with the current built in web server. Emhttp wasn't designed with multiple consecutive http requests in mind.

Sound familiar... I reached the same limit with my "awk" based server in unMENU.

We even have to parse GET ourselves (POST won't work) to get anything to work the way it does.

That is a pretty big issue.  Never gave much thought to it how unRAID submitted its pages.  I thought it did use POST

<form name="arrayOps" method="POST" action="/update.htm" target="progressFrame">

and it appears it does... What is it about POST that does not work for plugins?

I've yet to see conclusive evidence that simpleFeatures is causing crashes (on our setups we've never had a single issue, with up times reaching months). The majority of users seeing these problems are also running other plugins such as SABnzbd.

I've seen more issues with ulimit and lack of "low" ram than anything else.

It is rumored that Tom will be including lighttpd in unRAID 5.1, to help incorporate a plugin manager - and to help out plugin developers such as ourselves. At that point we can make a conscious effort to remove legacy code to provide a rock solid environment.

I remember him saying that too.  At that point I might expect an API to interface to his proprietary logic,  and a separation of it from the actual lighttpd web-server and presentation layers.

[speeding_ant]

That is a pretty big issue.  Never gave much thought to it how unRAID submitted its pages.  I thought it did use POST

 

I tell porkies. We couldn't use PHP sessions, not POST (need coffee). We do have to parse POST and GET values ourselves though.

 

I've seen more issues with ulimit and lack of "low" ram than anything else.

 

Yes. As PHP/emhttp *should* dealloc after each request there should be no issues with simpleFeatures - no matter how many http requests hit the server.

 

I remember him saying that too.  At that point I might expect an API to interface to his proprietary logic,  and a separation of it from the actual lighttpd web-server and presentation layers.

 

This would be ideal. That would give users the ability to easily redesign the web interface - not that it should be hacked together (Toms interface has been very well thought through, at a logic level anyway). At that point if you were interested in working on this, I believe working together would be the best option there.

[tw332]

Hoping for some help - I'm trying to research an issue but my home internet is down and the works connection is heavily firewalled - so please accept my apologies if this has been discussed previously.

 

I manually ran a routine "no correct" paritycheck from SF 1.0.11 (on RC11) overnight on Wednesday.  On returning home, my syslog showed >1500 (or >15000, forgive my poor memory) parity errors but also showed a line stating "check CORRECT".

 

Trying to debug, I ran another "no correct" parity check from SF.  When the page refreshed, the status changed to show the check is in progress but with the "write corrections" box ticked.  Checking the syslog again shows the "check CORRECT" line.

 

The parity errors I received are a seperate issue (which I'll deal with in another post if I can't resolve) but I'm trying to confirm if the line in the syslog is confirming that the parity disk was updated to reflect the changes or if the check carried out was actually a "NOCORRECT".  The results of the 2nd parity check returned 0 errors - so I'm unsure now whether parity was updated on the 1st check or if the errors showing were due to another issue.

 

If the parity disk was updated, is there something else I need to do to carry out a "NOCORRECT" parity check - both times I am positive I unticked the box to request a "NOCORRECT".

[badboyz]

No sensor temps for me:

 

Linux 3.4.26-unRAID.

root@RobertNas:~# sensors -u

k10temp-pci-00c3

Adapter: PCI adapter

temp1:

  temp1_input: 43.500

  temp1_max: 70.000

  temp1_crit: 90.000

  temp1_crit_hyst: 87.000

 

Motherboard/cpu :

ASRock E350M1 AMD E-350 APU (1.6GHz, Dual-Core) AMD A50M Hudson M1 Mini ITX Motherboard/CPU Combo

[trurl]

Temps were working for me with Atom CPU but not with upgrade to i3:

Linux 3.4.26-unRAID.
root@unSERVER:~# sensors -u
coretemp-isa-0000
Adapter: ISA adapter
Physical id 0:
  temp1_input: 35.000
  temp1_max: 85.000
  temp1_crit: 105.000
  temp1_crit_alarm: 0.000
Core 0:
  temp2_input: 33.000
  temp2_max: 85.000
  temp2_crit: 105.000
  temp2_crit_alarm: 0.000
Core 1:
  temp3_input: 36.000
  temp3_max: 85.000
  temp3_crit: 105.000
  temp3_crit_alarm: 0.000

[dmacias]

You need a custom sensors.conf for your motherboard. You can google your MB and sensors.conf and someone probably has made one.  I have an ASUS C60M1-I with an ITE IT8771E.  If you run sensors-detect there is no driver for IT8771E.  If I execute"modprobe it87 force_id=0x8721" I force it to use IT8721 which is supported and then the sensors show up.  But you need a custom sensors.conf to format the output correctly.  So I put this in my go script.  It copies my custom conf to /etc/sensors.d and then forces its use.

#Update Sensors
cp /boot/custom/etc/sensors.d/sensors.conf /etc/sensors.d
modprobe it87 force_id=0x8721

 

And here is my sensors.conf:

 

# lm-sensors config for Asus C60M1-I board

chip "it8721-*"

   label  in0   "Vcore"
   ignore in1
   label  in2   "+12V"
   ignore in4
   label  in5   "+5V"
   ignore in6

   compute  in2  @ * (50/12), @ / (50/12)
   compute  in5  @ * (205/120), @ / (205/120)
   compute  temp1 @ + 10, @ - 10

   label fan1 "CPU Fan"
   label fan2 "Chassis Fan"

   label temp1 "CPU Temp"
   label temp2 "MB Temp"
   ignore temp3

   set in0_min   0.5
   set in0_max   1.5
   set in2_min  12   * 0.95
   set in2_max  12   * 1.05
   set in3_min   3.3 * 0.95
   set in3_max   3.3 * 1.05
   set in5_min   5   * 0.95
   set in5_max   5   * 1.95
   set in7_min   3.3 * 0.95
   set in7_max   3.3 * 1.05

   set temp1_min 0
   set temp1_max 70
   set temp2_min 0
   set temp2_max 60

   set fan1_min 0
   set fan2_min 0

chip "k10temp-*"

   label temp1 "CPU Temp"

 

Also running sensors at command prompt outputs this:

 

k10temp-pci-00c3
Adapter: PCI adapter
CPU Temp:     +50.4 C  (high = +70.0 C)
                       (crit = +100.0 C, hyst = +97.0 C)

it8721-isa-0290
Adapter: ISA adapter
Vcore:        +1.09 V  (min =  +0.50 V, max =  +1.50 V)
+12V:        +12.20 V  (min = +11.40 V, max = +12.60 V)
+3.3V:        +3.38 V  (min =  +3.14 V, max =  +3.46 V)
+5V:          +5.00 V  (min =  +4.76 V, max =  +5.23 V)
3VSB:         +3.36 V  (min =  +3.14 V, max =  +3.46 V)
Vbat:         +3.36 V
CPU Fan:     2163 RPM  (min =    0 RPM)
Chassis Fan:  736 RPM  (min =    0 RPM)
CPU Temp:     +54.0 C  (low  =  +0.0 C, high = +70.0 C)  sensor = thermistor
MB Temp:      +40.0 C  (low  =  +0.0 C, high = +60.0 C)  sensor = thermistor
intrusion0:  ALARM

 

It would be nice to have this parsed and formatted under System Profiler like the other info there.  I have previously hacked it to show the raw output but a formatted sensors output would be nicer.

[Automatic]

Could someone please explain the feature "Enable background polling for spun-down disks"?  I've searched the forums and no luck.  What impact does it have on the drives?  What information does it provide?  Does it just run periodic SMART tests when the disk are spun down?

 

Thanks in advance

A SMART test may spin-up a disk when it is spun-down state this is usually the case for PATA connected drives, most SATA drives work fine. Therefore by default tests are disabled to prevent unnecessary spin-up of disks.

 

You can safely enable the setting and see if spun-down disks stay down (polling happens every 10 seconds).

 

Can I question this again? What I've gathered is it just runs periodic smart tests when they're idle? If so, say it finds an issue with a disk, what will it do?

 

And, what smart test does it do? Does it just look at the smart records or does it do something? If it finds an error will it email or what (I have email notifications set up)?

[Orbi]

First let me thank you speeding_ant and bonienl for your hard work and making unraid more enjoyable to use. It is a huge improvement over the stock interface and I've been using simplefeatures since version 1.0.5 without any single issue.

 

When version 1.0.11 was released, I upgraded unraid from rc8.0 to rc11 and simplefeatures from 1.0.5 to 1.0.11. All went well for over one week, rock-stable and responsive. You'll see in the syslog that on 3 march I powered down the server to include one more hard drive. The 3TB drive was preclearing for 3 days and on 6 march I stopped the array to include the drive and format it. The webgui never came up again and stayed unresponsive. Not knowing what to do I restarted the server. The webgui came up and for about 30 seconds I could browse around till it became unresponsive again. You'll see in the log that I reinstalled simplefeatures several times, but that did not solve the issue.

 

I formatted the new drive by restarting unraid with stock interface. After the drive was formatted, I restarted unraid with simplefeatures. It has been running since 6 march till this morning without problems. When I checked the webgui this morning, it was unresponsive. Everything else works, samba, mover, etc.

 

Syslog is attached. Thanks for the advice.

[optiman]

Anyone out there have a working sensors.conf for my setup, see below?

[dmacias]

Anyone out there have a working sensors.conf for my setup, see below?

 

I found this for your chip.  Dont know if it will be accurate.  What does sensors output show now.

 

#
# sample configuration for the Fintek f71882fg for MSI X58 Pro (7522-040R,
# like seen in Hetzner's EQ6 servers). To load the driver module, append a
# single line (without the #) to /etc/modules:
# f71882fg
#
chip "f71882fg-*"

# never change the in0, in7 and in8 compute, these are hardwired in the chip!
compute in0 (@ * 2), (@ / 2)
compute in2 (@ * 2), (@ / 2)
compute in3 (@ * 2), (@ / 2)
compute in4 (@ * 5.25), (@ / 5.25)
compute in5 (@ * 11), (@ / 11)
compute in6 (@ * 5.25), (@ / 5.25)
compute in7 (@ * 2), (@ / 2)
compute in8 (@ * 2), (@ / 2)

# Temperature
# cpu temp is about 35 deg celsius (linux, idle)
label temp1 "CPU"
set temp1_max 70
set temp1_crit 80
# the IOH runs rather hot, even with an idle system 60..65 deg celsius
# are seen. DO NOT ACTIVATE CPU SMART FAN OR IT WILL GET EVEN HOTTER!
label temp2 "IOH"
set temp2_max 80
set temp2_crit 90
# system is about 40 deg celsius (linux, idle)
label temp3 "System"
set temp3_max 50
set temp3_crit 55

# Fans
# value (CPU smart fan disabled!) is about 2300 rpm (linux, idle)
label fan1 "CPU"
ignore fan2
# value is about 2200 rpm (linux, idle)
label fan3 "System2"
ignore fan4

# Voltage
label in0 "3.3V"
label in1 "Vcore"
# BIOS (loaded) in1 == 1.25V
# linux idle in1 == 1.05V (?)
set in1_max 1.35
ignore in2
ignore in3
label in4 "+5V"
label in5 "12V"
ignore in6
label in7 "3VSB"
label in8 "Battery"