database error, all links are gone

today, i collected some links, in case that something bad happened screw my collection, i quit JD, make sure JD save the links, and restart it. unfortunately, bad things do happened. JD started, but there's nothing in windows. all my links are GONE? Why?

it has screwed me once. see "JD stucked, How to save my unfinished links?"

and IT SCREW ME AGAIN, IN A WEEK??? AAAAAAAAHHHH.................
i'm TOTALLY FREAKING OUT!!!:outch::outch::outch::outch::outch:

what's wrong with JD?
my question is :

1. i remain links after download for a record in case of reduplicate,
for now, it has been almost 10000 links saved in JD.
So is there some sort of limit of number of links saved in JD? like cannot over 10000?

2. how to make JD auto-backup its database?

3. how to rescue my links from this totally disaster:outch:


To Jiaz:
i have send my database.rar links to your massage box, please check and help me, thank you very much.

1) I keep my links in text files. jD isn't made for storing a long history of links. Links in jD consume a lot of memory. Also saving them becomes difficult after some time.

2) jD auto-saves its database every 10 minutes. If you would run a real-time/continuous backup program that can handle open files, you can backup the database as often as you want.

3) Wait for Jiaz, the saviour of databases.

Other related info :-

- There is a versioning option in jD that makes versions of your configuration and its links. Select File -> Save -> Backup or hit Ctrl-B whenever you add links to jD and this quagmire won't happen again.

- Note that the links in the Linkgrabber aren't saved. This is an expensive mistake that many customers make.

[Jiaz]

jd is not limited in any links counting. i saw ppl with 900k links in it *freaks*
it is important to give java enough memory to work with, we limit it by default to 512mb.
so with more links you need more memory

you must start jdownloader with
java -Xmx2048m -jar JDownloader.jar
in order to allow 2gb of ram usage.

next major version will use a lot less ram/cpu. easily can handle 10k links and more

try to start jd with more memory and it should be able to restore them (use the backup in case it does not work)

do not missuse jdownloader downloadlist as history!

TO remi:
thank you for your advise.

jD auto-saves its database every 10 minutes.

it seems not work that way. and i cannot find any config about auto-saves.

- Note that the links in the Linkgrabber aren't saved. This is an expensive mistake that many customers make.

i was aware of that kind of mistake month ago. it's not that problem.

maybe save the database manually is a good idea.

thx.




TO jiaz:

you must start jdownloader with
java -Xmx2048m -jar JDownloader.jar
in order to allow 2gb of ram usage.

may i ask:
how to exactly excute "start jdownloader with java -Xmx2048m -jar JDownloader.jar"
can you give me more details? i run the command in DOS, is that right?

i ran the command in DOS on my computer (2G memory)
it returns:

Error occurred during initialization of VM
Could not reserve enough space for object heap
Could not create the Java virtual machine.

and i borrowed a computer from a friend (3G memory)
and try it again, it returns the same : "Error occurred ......"

i changed the command to"java -Xmx1024m -jar JDownloader.jar"
but it seems still doesn't work. only returns "key not found...." some thing like that.

so i'm almost Desperate.
is there any chance to save my links back?
looking forward to your reply.
thx.

[editestowy]

Originally Posted by jdstuck

i changed the command to"java -Xmx-1024m -jar JDownloader.jar"

Correct command:
"java -Xmx1024m -jar JDownloader.jar"

But for 2GB memory machine I suggest to use default 512MB for jD (so use simple jdownloader.exe because it calls java with
-Xmx512m parameter). 1024MB seems to be too much and your comp will start to swap memory to HDD.

If you want jD to make automatic backup of the links (and options) - you may try to use Nightly version with Scheduler extension - it's possible to specify how often backup should be executed.

Correct command:
"java -Xmx1024m -jar JDownloader.jar"

sorry, it's a typing mistake in reply, not in DOS.

[Jiaz]

you must use at least 1024m, maybe 1500m
because many links need more ram in old stable.

i've let it go.

when the "next major version" release?

Then your jD will be stuck for some time but what would you expect with such an alias? Please read "More realistic project plan & schedule".

I have the same problem, when open the JD all the links dissappeared.

The only solution is wait for and update?

You are probably using JRE 1.6 update 26. You need to revert to a stable 32 bits version of the JRE 1.6. Update 17 or 21 are good ones. See "**External links are only visible to Support Staff**.

See also "Attention very important when upgrading to Java 1.6.0_26".

I'm using java 1.6.0_22 and having the same issue, the database is 16mb before I start jdownloader. After the start, it expands to 100MBs(now gonne to 400MBs, i will delete it and reinstall de old 16MB, no data lost for now). I tried to restore using "java -Xmx1024m -Dsun.java2d.d3d=false -jar JDownloader.jar" but after the program starts no files on list...

Using 1.6.0_22 since 10/2010, never had any problem...
Windows 7 x64

tip: To use the command you must enable the path to java.exe, try path="c:\(program folder to java)\jre6\bin"

Thanks in advance.

Update 22 is not the best version of java. I would remove it and replace it by JRE (1.)6 update 21. See "**External links are only visible to Support Staff**. Also use JavaRa before installing the new version to be sure that all remaining bits of java are removed.

Look likes this last update fixed this problem, at least to me.

Thank you very much for the help.

so how was this fixed? I have java version 6 update 17 (64 bit) and no links and an old backup database. I need a step by step guide, really not very computer savvy :( thanks in advance

If you want to know how to use the backup version, then just extract the files in the archive over the \config sub-folder that you'll find in your jD program folder.

If you want to know how to extract files from an archive, please consult the manual for your file archiving/compression tool.

BTW, you don't need a 64 bits version of java for running jD. A 32 bits version is sufficient for most applications.

I've extracted the backup zip to the config folder but it still is empty. the database.scrpit file is 543MB so it should have something in it but still no luck

In order to open such a large file you need to increase the memory space for running jD. Read post #3. Maybe a parameter -Xmx1024m is enough.

Here are two reasons for not storing too many links in jD :-

1) Don't use jD for keeping a history of links (yet). There are functions in jD to remove the completed/dead links. Keep your links in a separate text file.

2) Also adding links that will be downloaded next week is risky because links can go off-line and if they belong to a set you need to find mirror links or you're wasting time & bandwidth.

to change the memory parameter for java i followed the instructions on **External links are only visible to Support Staff****External links are only visible to Support Staff** and changed it to 1024, 2048 and 3072 but still no luck. i have 4GB or RAM so it should be able to handle it. let me know if thats how to change the parameters. thanks

You don't need to increase the java applet memory in your browser but the memory for starting up jD. Please, read post #3 again.

ok i've changed the java applet memory back to default and typed java -Xmx2048m -jar JDownloader.jar into my command prompt but all it says is 'Unable to access jarfile JDownloader.jar' What am I doing wrong?

[tony2long]

Open notepad, write that command, save it into JDownloader folder as something.bat and click it.

well it opened JDownloader but still no links. any other suggestions?

[tony2long]

If you want me to try fix it, see your PM. Or try older backup that is much smaller.

I used an old backup and only had a few links to recover. thanks for all the help though

[tony2long]

Thanks for the feedback.