Telnet time out

[GerryO]

Hi
I should start by saying I am not a technical person and networks leave me baffled but I can follow instructions.
As a bit of background I have two Humax machines owned both from new (2013) and, up until a couple of months ago, had no trouble, both are running custom firmware.
One of the machines has developed the Delete loop issue, it fails to delete selected files , will not copy over to external drive, titles get muddled with descriptions etc.
I have been using the Webif to delete files but copying to external USB was not working and I noticed that delted files would re-appear.
I have been putting off running the maintenance routines until yesterday when I decided to have a go. Everything seemed to be going smoothly and the Fixdisk was sorting things. It then started to deal with the problem I suspect and slowed to glacial.
I used my android tablet to run the Webif and from that I selected Telnet.
The first question I have is because the fix is taking so long the Telnet ends the sesssion and I have to reconnect, does this mean that Fixdisk has paused too or is the routine still running on the Humax?
Second question, If I were to decide not to continue with the routine and select end from telnet menu how would that leave the Humax drive? To make that a bit clearer, would it brick the drive, cause errors or retrun to previous state.

 

[Black Hole]

I have been putting off running the maintenance routines until yesterday

That's a mistake. Should have been at the first sign of trouble.

The first question I have is because the fix is taking so long the Telnet ends the sesssion and I have to reconnect, does this mean that Fixdisk has paused too or is the routine still running on the Humax?

No. If you fire up Telnet again, you can reconnect to the existing session. It can take days if there is a real mess to sort out.

Second question, If I were to decide not to continue with the routine and select end from telnet menu how would that leave the Humax drive? To make that a bit clearer, would it brick the drive, cause errors or retrun to previous state.

Not advisable unless you have to, but hopefully a re-run of fixdisk would fix it.

 

[everthewatcher]

That's a mistake. Should have been at the first sign of trouble.

I've got into the habit of running fdisk at the end of the month.

 

[GerryO]

Hi
I should start by saying I am not a technical person and networks leave me baffled but I can follow instructions.
As a bit of background I have two Humax machines owned both from new (2013) and, up until a couple of months ago, had no trouble, both are running custom firmware.
One of the machines has developed the Delete loop issue, it fails to delete selected files , will not copy over to external drive, titles get muddled with descriptions etc.
I have been using the Webif to delete files but copying to external USB was not working and I noticed that delted files would re-appear.
I have been putting off running the maintenance routines until yesterday when I decided to have a go. Everything seemed to be going smoothly and the Fixdisk was sorting things. It then started to deal with the problem I suspect and slowed to glacial.
I used my android tablet to run the Webif and from that I selected Telnet.
The first question I have is because the fix is taking so long the Telnet ends the sesssion and I have to reconnect, does this mean that Fixdisk has paused too or is the routine still running on the Humax?
Second question, If I were to decide not to continue with the routine and select end from telnet menu how would that leave the Humax drive? To make that a bit clearer, would it brick the drive, cause errors or retrun to previous state.

Thanks for the fast replies.
Black Hole, I am seriously technophobic so I was quite surprised how everything worked out as stated in the forums advice posts.I shall leave fixdisc running.
Supplementary question, does fixdisk clone every file or just the ones it detects errors in?

 

[prpr]

It doesn't clone anything. It fixes (or at least tries to) irregularities in the internal filesystem management structures. Sometimes this may mean files get corrupted or truncated or deleted, but generally it doesn't. Depending on the degree of damage, it can take a non-predictable and non-trivial amount of time to do so. Sometimes it's quicker to backup, reformat and restore, but you generally only discover this with hindsight.

 

[Black Hole]

I was quite surprised how everything worked out as stated in the forums advice posts

:sigh: If you're that technophobic, maybe you should avoid technology?

This ain't just any forum!

 

[GerryO]

It doesn't clone anything. It fixes (or at least tries to) irregularities in the internal filesystem management structures. Sometimes this may mean files get corrupted or truncated or deleted, but generally it doesn't. Depending on the degree of damage, it can take a non-predictable and non-trivial amount of time to do so. Sometimes it's quicker to backup, reformat and restore, but you generally only discover this with hindsight.

Ahh, Thanks, I used the term clone because thats what telnet shows, I had wondered.
It now seems to be getting through at a better speed. I did read a post on here where it took something like ten days!
The other problem it had was it wouldnt copy any of the files to the external drive so I figured this would be a less complicated solution.

 

[GerryO]

:sigh: If you're that technophobic, maybe you should avoid technology?

This ain't just any forum!

Yes, well in answer to that its almost Impossible to avoid technology these days in any case one cannot let a phobia rule ones life, they need to be confronted to get past them.
I chose this forum to post the question as it had the most relevant title Customised Firmware.
I had read the posts in the link you gave but could find nothing in there that addressed the time out question.
Thanks to you guys on here I got the answer.

 

[Black Hole]

I used the term clone because thats what telnet shows

Really?

I chose this forum to post the question as it had the most relevant title Customised Firmware.

Of course. Where else?

I had read the posts in the link you gave but could find nothing in there that addressed the time out question.

Not in so many words perhaps, but:

You do not need to remain connected while the process runs (so long as you gave it the "-y" option).

 

[GerryO]

Really?


Of course. Where else?


Not in so many words perhaps, but:

Yes it says: "Clone multiply-claimed blocks? Yes".

Not in so many words for sure but It also says not use -y unless advised by a knowledgeable forum member, so I didn't, hence the question.

 

[Black Hole]

It also says not use -y unless advised by a knowledgeable forum member

With the Telnet menu (as per Step 2), select the fix disk option by typing the relevant command from the menu (currently "fixdisk"), reply "-y" to "additional options" (saves you having to type "y" each time there's a query), and let it run. The process may take from a few minutes to several hours (or even days). You do not need to remain connected while the process runs (so long as you gave it the "-y" option).

 

[/df]

When fix-disk, or actually e2fsck, uses the word "clone", that's just a part of the fixing.

Sometimes bad things happen that can lead to the filesystem believing that two different files both use some bit of the disk ("multiply-claimed blocks"). This has to be wrong, but the fixing process doesn't know which of the files is the rightful occupier of the blocks (or indeed, if either of the files is). Its best attempt to solve the problem is to find a matching number of blocks that it believes are free and copy ("clone") the data from the multiply-claimed blocks to the free blocks; then the indexes for one of the claimant files are updated to replace each reference to a multiply-claimed block with a reference to the corresponding newly allocated block, which of course is no longer in the free list. So there you have it.

 

[GerryO]

hWhen fix-disk, or actually e2fsck, uses the word "clone", that's just a part of the fixing.

Sometimes bad things happen that can lead to the filesystem believing that two different files both use some bit of the disk ("multiply-claimed blocks"). This has to be wrong, but the fixing process doesn't know which of the files is the rightful occupier of the blocks (or indeed, if either of the files is). Its best attempt to solve the problem is to find a matching number of blocks that it believes are free and copy ("clone") the data from the multiply-claimed blocks to the free blocks; then the indexes for one of the claimant files are updated to replace each reference to a multiply-claimed block with a reference to the corresponding newly allocated block, which of course is no longer in the free list. So there you have it.

Thank you for explaining that, I have a better understanding now.
The routine is still running. The drive size is 500 g and there was, if I remember correctly, just 15g of space left. I think I am in for a long wait. Lesson learnt about early intervention.

 

[GerryO]

Well it has done the job. It finished at today, Exit status 0.
All seems good now much snappier too.
I did notice in the report some files were dated 2011, there were no files that old being kept by me.
Should I run the fixdisc routine every few months or so to keep things tidy or is there another routine better suited?

 

[MartinLiddle]

Well it has done the job. It finished at today, Exit status 0.
All seems good now much snappier too.

It might be helpful if you posted the hard drive SMART data which can be accessed via the webif under Diagnostics> disk diagnostics. It will allow us to advise you of the health of the hard drive.

 

[GerryO]

Thanks, I will give it a try.