Vix 3.0 ImageManager Crashes at 70% and box needs reflash

[sledge121]

ImageManager crashes at 70% saying device busy, box stays on but drops telnet and ssh requests, when box is rebooted it doesn't start and needs reflash, it's happeded twice now, don't want to go for a third without a fix. Should I roll back to 2.4, is ImageManger more stable in that version.

[Larry-G]

ImageManager crashes at 70% saying device busy, box stays on but drops telnet and ssh requests, when box is rebooted it doesn't start and needs reflash, it's happeded twice now, don't want to go for a third without a fix. Should I roll back to 2.4, is ImageManger more stable in that version.


Which exact build(s) have you had this problem with, so that we can at least try to replicate your problem ?. also for the record i have made approximately 100 full image and settings backups from within the 3.0 image series on 3 different receivers and all work fine for me.

please provide us with as much information as you can such as which builds this is happening with and where the backups are being created such as a internal HDD or a mounted USB stick etc. also are you doing any thing else on the receiver while the backups are being created ? again provide as much info as you can so that we can start to look at where the problem may lie.

[Rob van der Does]

Debuglogs are essential to find a problem.
And indeed making and restoring backup (both settings & image) are an integral part of our beta-testing; sofar we found no problems in this area. So if you do: logs please!

[dfdream]

All I was making a backup on my SOLO yesterday evening before updating from 242 to 250 (vix 3.0) and the exact same thing happened.
It was running in the background so I dont know at what stage it failed. I backup to HDD on my DUO (done that 100s of times). The box froze and when I powered it off the box didnt do anything.. I reflashed with 204 backup I had and applied updates to 250 and did new backup and it worked 100%.
Thought I might have backed up to flash but the temp folders are there on the DUO where they should be.
Will check later if there are any logs I can send...


ImageManager crashes at 70% saying device busy, box stays on but drops telnet and ssh requests, when box is rebooted it doesn't start and needs reflash, it's happeded twice now, don't want to go for a third without a fix. Should I roll back to 2.4, is ImageManger more stable in that version.

[sledge121]

Hi guys, thanks for the responses, I am using the following

Openvix 3.0.245 vusolo
CCcam 2.3.0 (connection up)
OpenVPN using PKI certs (tunnel up)
CrossEPG with EPG stored on USB
Sandisk fit 8GB (loads of space)

All config and settings work as expected, then I go into ImageManager and select new backup, it hangs around for an extended period at 40% then jumps to 70% and an error box appears on the screen, the error says something about device busy, when I press ok ImageManager says failed and stays at 70%. At this stage I can still see and hear the TV programme but cannot telnet/SSH to the box or change channel, after a reboot the box is hosed and needs a reflash.
At first I thought it was a USB stick issue so I tried a new sandisk cruzer blade 8GB but exactly the same happened, any advice is welcome.

Thanks
Sledge

[dfdream]

Actually Sledge I think mine was 245 as well.
As I had a backup from 3.0 204 I flashed with that and did online update to 250.
The imagemanager worked fine after that.
Try a different version of 3.0 and do online updates (you need to turn off stable updates to in Online Updates Settings ) to 250.
Should be OK after that. Or maybe do online updates to 250 with 245 to see if that solves it...


Hi guys, thanks for the responses, I am using the following

Openvix 3.0.245 vusolo
CCcam 2.3.0 (connection up)
OpenVPN using PKI certs (tunnel up)
CrossEPG with EPG stored on USB
Sandisk fit 8GB (loads of space)

All config and settings work as expected, then I go into ImageManager and select new backup, it hangs around for an extended period at 40% then jumps to 70% and an error box appears on the screen, the error says something about device busy, when I press ok ImageManager says failed and stays at 70%. At this stage I can still see and hear the TV programme but cannot telnet/SSH to the box or change channel, after a reboot the box is hosed and needs a reflash.
At first I thought it was a USB stick issue so I tried a new sandisk cruzer blade 8GB but exactly the same happened, any advice is welcome.

Thanks
Sledge

[sledge121]

I like the fact that it doesn't happen in 250 but I dont like the idea of 250 being an unstable release.

[dfdream]

Maybe go back to 242 or earlier...
250 has been going as good as 245 or 242 for me anyway on the SOLO...


I like the fact that it doesn't happen in 250 but I dont like the idea of 250 being an unstable release.

[sledge121]

dfdream, did you have the exact issue I was having.

[Larry-G]

I like the fact that it doesn't happen in 250 but I dont like the idea of 250 being an unstable release.

I just did a full image backup on my ultimo build 255 ( i recently cleared out my backups but i also have backups from builds 204 and 249 ) directly to a 8GB sandisk cruzor blade memory stick and it all went fine, it is perfectly ok and expected to hold a while on 40%. also make sure that while your creating the backup your not doing any thing at all on the receiver, dont even touch the remote.

i'm just about to update this box to the latest build then i'll test a image backup on that one too.

as for the image being "unstable". just because its set as unstable does not necessarily mean some thing is wrong with it, we set images as unstable for many reasons, some as simple as we may not have tested a certain bug fix prior to release.

[dfdream]

Only once...
Flashed back with 204 backup and did online updates to 250 and backup ran fine.
Dont know if its just a 245 issue.

D

[sledge121]

I just reflashed the box, stuck my config on, upgraded to 250 and the crash happened again, see error below (I typed by hand from the on screen error as the box is now hosed and needs a reflash, well, when I reboot it)

imagemanager
error:[failure instance:traceback:<type
'exceptions.OSerror'>[erno 16]device or resource busy:
'/mediad/hdd/imagebackups/vusolo-mount/root/dev'
/usr/lib/python2.7/threading.py:525:_bootstrap
/usr/lib/python2.7/threading.py:552:_bootstrap_inner
/usr/lib/python2.7/threading.py:505:run
---<exceptioncaught here>---
/usr/lib/python2.7/site-
packages/twisted/python/threadpool.py:207:_worker
/usr/lib/python2.7/site-
packages/twisted/python/context.py:118:callwithcontent
/usr/lib/python2.7/site-
packages/twisted/python/context.py:81:callwithcontent
/usr/lib/enigma2/python/Plugins/systemplugins/vix/imagemanager.py:1019:dobackup3
/usr/lib/python2.7/shutil.py:244:rmtree
/usr/lib/python2.7/shutil.py:244:rmtree
/usr/lib/python2.7/shutil.py:253:rmtree
/usr/lib/python2.7/shutil.py:251:rmtree
]

[judge]

Not sure what you mean by 'stuck my config on'?
If there is a crash, there will be a crash log created in /home/root/logs/
Upload that to save you typing what's displayed on screen.

[Larry-G]

also try setting your USB as the default logs location so that if you have to reflash you dont loose any of those precious logs.

[sledge121]

By stuck my config on I mean Openvpn, bouquets, picons, satellites file, cron, CCcam etc, the imagemanager is so that I have a full backup of my system and to save me time provisioning another box of the same type. It takes me about 45 mins to configure it from scratch, I will have another go and change the log location as you say, after the imagemanager fails it wont allow any access and after a reboot it won't start. Does the above error shed any light on whats happening.

[pooface]

By stuck my config on I mean Openvpn, bouquets, picons, satellites file, cron, CCcam etc, the imagemanager is so that I have a full backup of my system and to save me time provisioning another box of the same type. It takes me about 45 mins to configure it from scratch, I will have another go and change the log location as you say, after the imagemanager fails it wont allow any access and after a reboot it won't start. Does the above error shed any light on whats happening.

Another thing to try, is that if you have a null modem cable, you can connect the box via that and create a serial log to show exactly what's happening etc, and that would hopefully solve the problem better...

[belierzz]

Could it be the Solo is running out of memory? Do you have a swap enabled? Else the flash might fill. I experienced this on a Solo running another image... With other boxes like Uno / Ultimo this shouldn't happen.

[pooface]

afaik, the image creates a swap file if needed?!

[belierzz]

Jep, Vix does. But the symptoms look the same... What happens if the stick is e.g. in FAT32?

[Rob van der Does]

Only one solution: debuglogs!

[sledge121]

Thanks guys, what is the process to create a swap, I will do it and have another pop.

And I will send logs to usb, so long as it doesn't compund the issue.

[judge]

Menu -> Set up -> VIX -> Swap Manager.

[sledge121]

Update

Vix 3.0 250
USB ext3 with maximum swap size enabled
Image manager successfully backs up

This was not a true test of my original configuration as I did not start any extra daemons this time I.e. cron, cccam, openvpn, the next test will be to enable these and attempt another image backup, if this is successful then it must have been running out of memory as previously suggested and the addition of the swap fixed it.

My next questions are,
I now have a swap on an ext3 partition and have created an image of this configuration, if I configure another solo by flashing with this image I presume the USB that i add after for epg and picons etc will need to be pre formatted as ext3?

Now that swap will be running on USB, what will happen if the USB is removed when the box is running.

Thanks for all your help, I have donated to this excellent site.

Sledge

[sledge121]

Problem solved

After hours of my life lost forever I discovered the crash doesn't happen if you format the USB as ext3, mine was fat32. If I had done that from the start it would never of had a problem, I don't even need to enable swap now unless anybody recommends it for other reasons.

Thanks for all the help though, it's helped me understand things better.

Thanks
Sledge

[Larry-G]

Problem solved

After hours of my life lost forever I discovered the crash doesn't happen if you format the USB as ext3, mine was fat32. If I had done that from the start it would never of had a problem, I don't even need to enable swap now unless anybody recommends it for other reasons.

Thanks for all the help though, it's helped me understand things better.

Thanks
Sledge

all my 3 receivers use fat32 format usb sticks and i never had any problems like yours.

[sledge121]

I don't understand what was happening then, are your boxes vu solo. I had to covert the USB to ext3 to enable swap then I noticed a previous comment that the os will create swap when it needs to but only if USB is ext3, so I disabled swap, rebooted, ran image manger and it worked.

[Larry-G]

sorry no, i dont have the solo. got every other vu though.

any hoo if it works for you great.