Is Twonky gone? Cant restart

I have a MyBookLive 1TB, and TwonkyMediaServer was working fine, and then last night it stopped. I cannot gain access to the Twonky UI at port 9000, nor does disabling and enable the Twonky service in the MyBookLive UI accomplish anything. I have tried this over and over again, powered down the drive, my routers, Mac, everything to try to gain access again. My Xbox360 no longer sees the service obviously either.

The only thing I changed in the Twonky UI prior to this strange error, was to have the Xbox view it by Folder by default. It worked for a few minutes, then after that, it stopped working and cannot gain access to GUI.

This is a huge part of my home network, how can I get it back working ASAP?

I even tried to use a old config backup from last month. A factory restore is not an option as I have backups I do not want to lose.

1 Like

Only fix was a factory restore, disappointed with that !

I am having the same issue. I have done MANY factory resets, recopied all of my files back to the drive(for hours) only to have access to twonkey disappear. How can I fix this?

one more thing. The reason I had to restart the server was do to it not rescanning new content. Default was -1 which is supposed to be constant scanning. I changed it to 10 minutes and had to restart twonky. Got the little arrow circle and please wait message but twonky NEVER came back, after hours I closed the browser window and now can’t access the UI. I am using a 64bit windows vista laptop and also have an Ubuntu  running 11.4 beta.

Could be that the TwonkyServer database is corrupted.  Erasing two files and then starting TwonkyServer.

My personal entry to my own technical knowledge base…

15:56 01/05/2011

For some reason TwonkyServer would not start.  Kept on starting and shutting down every time the media server was started.

To login as root on the MyBook Live you need a SSH client like the one at http://www.chiark.greenend.org.uk/~sgtatham/putty/ .

Solution…

Login to Dashboard and disable Twonky.

Enable SSH ( http://mybooklive/UI/ssh) and login as root using the SSL client.

Commands to issue at the Linux command prompt.

  • cd /CacheVolume/twonkymedia
  • mv twonky.db PossiblyCorrupt_twonky.db
  • mv db.info PossiblyCorrupt_db.info
  • logout

The reason the instructions is to rename (mv) the files and not erase them is that my mantra is to keep a copy of anything that’s changed. Just incase some tweak may need to be reversed at a future date. If it all works then PossiblyCorrupt_twonky.db and PossiblyCorrupt_db.info can be erased.

Login to Dashboard and enable Twonky.

Login to Twonky’s web interface  ( http://MyBookLive:9000/)

The status screen should show that the scan is in progress. There is no need to manually rebuild or refresh Twonky’s media database. The database has just been effectivley erased so Twonky has to create a fresh media database.

In response to the error…

Log:

15:36:11:024 **mediaserver starts.**
15:36:11:024 LOG_HTTP:upnp_http_register_handler registered a handler for /DeviceDescription.xml 3
15:36:11:024 LOG_HTTP:upnp_http_register_handler registered a handler for /ConnectionManager.xml 3
15:36:11:024 LOG_HTTP:upnp_http_register_handler registered a handler for /ConnectionManager/Control 4
15:36:11:024 LOG_HTTP:upnp_http_register_handler registered a handler for /ContentDirectory.xml 3
15:36:11:024 LOG_HTTP:upnp_http_register_handler registered a handler for /ContentDirectory/Control 4
15:36:11:024 LOG_HTTP:upnp_http_register_handler registered a handler for /MediaReceiverRegistrar.xml 3
15:36:11:024 LOG_HTTP:upnp_http_register_handler registered a handler for /MediaReceiverRegistrar/Control 4
15:36:11:024 LOG_HTTP:upnp_http_register_handler registered a handler for 8
15:36:11:024 LOG_HTTP:upnp_http_register_handler registered a handler for 16
15:36:11:024 LOG_HTTP:upnp_http_register_handler registered a handler for /disk 3
15:36:11:025 LOG_HTTP:upnp_http_register_handler registered a handler for /httpproxy 3
15:36:11:025 LOG_HTTP:upnp_http_register_handler registered a handler for /albumart 3
15:36:11:025 LOG_HTTP:upnp_http_register_handler registered a handler for /configpage 7
15:36:11:025 LOG_HTTP:upnp_http_register_handler registered a handler for /config 7
15:36:11:025 LOG_HTTP:upnp_http_register_handler registered a handler for /setup 7
15:36:11:025 LOG_HTTP:upnp_http_register_handler registered a handler for /images 7
15:36:11:025 LOG_HTTP:upnp_http_register_handler registered a handler for /index 7
15:36:11:025 LOG_HTTP:upnp_http_register_handler registered a handler for /resources 7
15:36:11:025 LOG_HTTP:upnp_http_register_handler registered a handler for /webbrowse 7
15:36:11:025 LOG_HTTP:upnp_http_register_handler registered a handler for /cgi-bin 7
15:36:11:025 LOG_HTTP:upnp_http_register_handler registered a handler for /bgtrans 3
15:36:11:025 LOG_HTTP:upnp_http_register_handler registered a handler for /createobject 7
15:36:11:025 LOG_HTTP:upnp_http_register_handler registered a handler for /rpc 5
15:36:11:025 LOG_HTTP:upnp_http_register_handler registered a handler for /rss 3
15:36:11:025 LOG_HTTP:upnp_http_register_handler registered a handler for /mediarss 3
15:36:11:025 LOG_HTTP:upnp_http_register_handler registered a handler for /wii/feed 3
15:36:11:026 LOG_HTTP:upnp_http_register_handler registered a handler for /json/feed/ 3
15:36:11:026 [Info] dbx_init - using database: /CacheVolume/twonkymedia/twonky.db
15:36:11:027 [Error] dbx_query - SQL error: database disk image is malformed
15:36:11:027 sqlError - I/O error occured (11). Checking database location
15:36:11:027 sqlError - regular database corrupted, invoking rebuild
15:36:11:027 LOG_SYSTEM:upnp_init_thread (id=1208349824) with name rpc_stop_server
15:36:11:727 LOG_SYSTEM:upnp_server_stop Server terminated
15:36:11:927 dbx_queue_release called
15:36:11:927 LOG_SYSTEM:file_unlink error (errno=2,filename=/CacheVolume/twonkymedia/twonkymedia-locations.db)
15:36:11:928 LOG_SYSTEM:file_unlink error (errno=2,filename=/CacheVolume/twonkymedia/5.1.9)
15:36:11:928 [Error] dbx_delete_db - called without dbx_open
15:36:11:928 LOG_SCAN: upnp_file_scanner_exit(): In
15:36:11:928 LOG_SCAN: upnp_file_scanner_lock - critical section not initialized
15:36:11:928 LOG_SCAN: upnp_file_scanner_unlock - critical section not initialized
15:36:11:928 LOG_SCAN: upnp_file_scanner_exit - critical section not initialized
15:36:11:928 LOG_SCAN: upnp_file_scanner_exit(): Out
15:36:11:928 LOG_SYSTEM:upnp_cds_actions_release()
15:36:11:929 [Error] dbx_close - called without dbx_open
15:36:11:929 **mediaserver stops.**
4 Likes

Thanks for this workaround but how can I reach the folder

/CacheVolume/twonkymedia

It does not seem to be there. I am using Mac OS X.6

Does anyone know this?

(The support team suggests to format the drive … great support!)

Ah, got it. And it worked! Thanks!!

Nice one that the solution worked.  It’s a good NAS drive but it is a pity that Western Digital’s support policy seems fails badly.

They are following Apple’s example to the looks of it. With Apple, their computers are now getting infected with fake anti-virus programs. Apples policy is if it is determined that this is the case then apart from directing the user a a few bonafide anti-virus companies, the user is NOT to be assisted.  The fix is actually quite easy and a simple google search will reveal the solution.

I’m now beginning to believe that Western Digital may be taking the same stance.  The fault could be stupidly simple, like in this case erasing two files and letting TwonkyMedia create the database, but no…  Western Digital’s answer is to wipe the lot to fix.  Throw the baby out with the bath water!!!

It’s a really sad situation.   :cry:

Just to help others find this…  Can you mark my post (the one with instructions) as the solution and also kudos it?  :smiley:

Prof_Hick wrote:

Ah, got it. And it worked! Thanks!!

1 Like

I kudoed it already but I don’t know how to mark it as the solution (which it is!)

Good point… Seems that option is not on the community?  Pass.  Don’t worry about it.  :slight_smile:

Main thing is that you’ve got Twonky working ansd (hope WDC support staff are reading this) there was no need to factory restore the MyBook Live and wipe it’s contents.  :robotvery-happy:

Dear Myron

   I’m a new user  and my first post.

Your simple solution worked as a charm.Great tip.I too faced this problem and was in the verge of giving up.Then saw this post and it worked.Twonky now retains its settings ,updates its database anatomically and its streams data across.

 However I do have a a couple of question.

A) When I purchased my drive and plugged in to system it was showing 2.7TB as against anticipated 3TB. Where is the rest of space ?Is it reclaimable or its marked for system’s use.

B) I always observed that if I move files within the share it happens pretty quickly ,but when I tried to move a big files across shares its taking so much of time.Why is this.?

Throwing some light there is greatly appreciated.

Nice day ahead

-Eff.

Depends on how it’s doing its math.

Pardon the long numbers, but most “MARKETING” terms means that 1 TB  is

1,000,000,000,000 bytes.

But computers calculate using Base 2 math, where a terabyte is actually

1,099,511,627,776 bytes.

So, converting, 3 TB Marketting to computer TB is 2.79.

TonyPh12345 wrote:

Depends on how it’s doing its math.

 

Pardon the long numbers, but most “MARKETING” terms means that 1 TB  is

 

1,000,000,000,000 bytes.

 

But computers calculate using Base 2 math, where a terabyte is actually

 

1,099,511,627,776 bytes.

 

So, converting, 3 TB Marketting to computer TB is 2.79.

 

 

Actually it’s MS who does this wrong - they do this in binary instead of 10base math which is essentially what you said above but puts the onus on MS to fix the issue in a future version of Windows (which people have been asking them to do for many versions now). Hard drive manufactuers correctly indicate that 1TB = 1,000,000,000,000 bytes. - It’s your operating system thats reading it incorrectly as 1,099,511,627,776 bytes. I think Mac OSX now properly reads a TB as 1,000,000,000,000 bytes.

There is likely a small ammount of space also being used to house the drives operating system, etc. which may be taking up a small fraction of the space on the drive- but the majority of it is not actually missing.

Well, I wouldn’t really say that Microsoft does it “wrong”.  Memory and all that has used K = 1024, pretty much since the beginning of the microchip, even before hard drives came along.

Part of the problem goes back to floppies… depending on the OS and the formatting used, the same number of physical data bits on the disc surface resulted in vastly different actual usable storage amounts.  A 3.5-inch DSDD disc was just sold as DSDD (double-sided, double-density) and no storage amount was given (unless the discs were pre-formatted).  Formatting the exact same disc under DOS, you’d get 720K… format it on a Mac (with less overhead) and you had 800K… format it on an Amiga and you had 880K available.

So, the media manufacturers have never wanted to be tied down to one operating system or one specific type of formatting.

The newer formats the OSes use don’t have nearly the discrepancies in the amount of overhead, so the differences in usable space aren’t that great depending on the formatting.

But there still is overhead, and it takes more space to store your stuff than what the stuff itself takes up, so there will always be a difference between the unformatted capacity and what is actually usable.

One can only assume that somewhere along the line, some bright person in marketing noticed that if they stick to the SI prefixes, as opposed to the conventional definitions, that their hard drives looked bigger… so all the manufacturers have been doing it for 30 years.  Doesn’t matter whether it was Mac, or Amiga, or Atari, or Win/DOS, or Silicon Graphics, or anything else… your formatted capacity never matched the drive label.  This “problem” isn’t new.

All apple has done, is to start using the SI prefixes instead of the standard computer conventions, for reporting size, so that the OS reports the same as what is on the manufacturer’s label.  But, by doing that a 1MB file no longer takes up 1MB of system memory – because the computer chips still use the conventional prefixes. :wink:

No manufacturer is about to go back to conventional prefixes, because then their drives will look smaller, on the store shelf, and folks will buy the competition that still marks using SI prefixes.  A 1TB (empty, SI prefix) looks bigger than a 931 GB drive (formatted, base-2 prefixes), even if they’re the exact same thing.

Well, the SI “owns” the definition of those prefixes…

TERA DOES mean 1,000,000,000,000, even though “history” and past convention to use it as 2^40 in computer contexts is, well, wrong and risky.

For everyone to be on the same page, they need to be using the IEC Standard term TEBI instead of TERA. 

The IEC standards should be more commonly adopted to prevent ambiguity and confusion.  A Tebibyte (abbreviated TiB) is 2^40, which is how “Progressive” OS’s are reporting numbers.   Linux is SLOWLY making its way there…

But either way, WD’s NUMBER is correct, when one considers the FOOTNOTE they put on the packages for hard drives… :wink:

Yeah… good luck with that one. I have a hard enough time explaining to people the difference between a Terabyte and a Gigabyte… dreading the thought of adding Tebi’s to the mix… You don’t know how many times I’ve heard people refer to things such as “Teragigs” “Gigatears” “Teramegs” and the like…

Well, right or wrong, conventions are hard to get away from.

Electricity is still usually shown, by convention, to flow from + to -, even though the electrons are really flowing from - to +, and the earth’s “north” magnetic pole is really the “south” magnetic pole, since magentic poles were defined as what points “north”, and opposites attract and likes repel.

So, since in the computer world the base-2 prefixes have been used, long before Microsoft was founded, I don’t see any changes, whether “official” or not, taking hold overnight.  I see being “stuck” with 1K=1024 for quite some time.

Thank you Myron!

Your solution worked perfectly for me to fix a broken Twonky Media Server.