[X] My Assistant

Helps if you post the error. I'm guessing you're trying to use port 80, which won't work with a non-privileged user.

Nope. Port 4xxxx

Error under a different user other than root.

2015-04-20T21:31:05Z [info] The database could not be loaded. Please check file permissions and file system integrity.
2015-04-20T21:31:05Z [info] If everything appears to be working, please do the following:
2015-04-20T21:31:05Z [info] 1. Locate the directory /mnt/pools/A/A0/NAS Primary/HentaiAtHome_1.2.5/data
2015-04-20T21:31:05Z [info] 2. Delete the file hath.db
2015-04-20T21:31:05Z [info] 3. Restart the client.
2015-04-20T21:31:05Z [info] The system should now rebuild the database.
2015-04-20T21:31:05Z [info] ***************************************************************************************************************
2015-04-20T21:31:05Z [info]
2015-04-20T21:31:05Z [ERROR] {org.hath.base.HentaiAtHomeClient.dieWithError(HentaiAtHomeClient.java:284)} Critical Error: Failed to load the database.
2015-04-20T21:31:05Z [info] Shutting down...
2015-04-20T21:31:05Z [info] Your business is appreciated
2015-04-20T21:31:05Z [info] Logging ended.

2015-04-20T21:31:05Z [info] 1. Locate the directory /mnt/pools/A/A0/NAS Primary/HentaiAtHome_1.2.5/data

Start by removing that space from the directory "NAS Primary"

Start by removing that space from the directory "NAS Primary"

Is it really a pathing issue? I've had it running under root with no problems.

Go investigate and post your results, as messages given by your machine. Stop believing that people can either read your mind or can connect to your machine and check.

Go investigate and post your results, as messages given by your machine. Stop believing that people can either read your mind or can connect to your machine and check.

I'm not. I just have doubts that it's actually a pathing issue. I don't understand the reasoning behind it as I have it running for hours on user root. I've been hovering at 900-1000 trust and my quality has been 6000+.

What reason would make it a pathing issue for other users other than root?

try

CODE

mkdir My Directory Like On MS Windows
cd My Directory Like On MS WIndows

and see what happens

Also, [catb.org] read this before you annoy the people that might be willing to help you.

try

CODE

mkdir My Directory Like On MS Windows
cd My Directory Like On MS WIndows

and see what happens

Makes a bunch of directories for every word.
Only goes to directory my when using second command.
This applies to every user. Root included. Yet HentaiAtHome runs normally only on root.
Just asking the reasoning behind any difference in pathing from root and other users before I kill the process and fuck up all my torrent locations.

I've been using backslashes as a workaround for this.

CODE

mkdir My\ Directory\ Like\ On\ MS\ Windows
cd My\ Directory\ Like\ On\ MS\ Windows

I didn't mean to be annoying. I want to understand this as much as you do.

Oh OK... you made me curious enough to make a testing "script", what is the output of:

CODE

diff <(find /mnt -type d -perm -111) <(find /mnt -type d -perm /111)
diff <(find /mnt -type d -perm -444) <(find /mnt -type d -perm /444)

?

This should catch directories with strange permissions. /mnt shan't have any sensitive files either, so it isn't a problem to post it.

Still, the engineering process happens by trial and error and not by a full understanding. In the majority of cases the solution is found before understanding the problem, and only after finding a solution people start asking the question of what actually happened. (to be fair there is an argument of the *BSD guys that the Linux guys do not perform that last part of the reasoning (IMG:[invalid] style_emoticons/default/smile.gif) )

Hi, just come back for my result

client 1: H@H client that seems to have "issues" in hathdl

client 2: for testing the hathdl download in another computer & different H@H client

after I keep client 1 running a day or two, it begin to download the hathdl, but there something different,

in client 2 when I drop the hathdl into the folder the download begin immediately, but in client 1 I think it need about 6 more hours after drop the hathdl to the folder to begin the download

so that why I think client 1 "not work" when I see its not downloading, maybe I'll replace all the setting only keep the cache to see is there any different when I have time

Thanks

Oh OK... you made me curious enough to make a testing "script", what is the output of:

CODE

diff <(find /mnt -type d -perm -111) <(find /mnt -type d -perm /111)
diff <(find /mnt -type d -perm -444) <(find /mnt -type d -perm /444)

?

This should catch directories with strange permissions. /mnt shan't have any sensitive files either, so it isn't a problem to post it.

-sh: syntax error near unexpected token `('

Also changed the directory name. Same results.

2015-04-21T07:50:39Z [info] The database could not be loaded. Please check file permissions and file system integrity.
2015-04-21T07:50:39Z [info] If everything appears to be working, please do the following:
2015-04-21T07:50:39Z [info] 1. Locate the directory /mnt/pools/A/A0/NASPrimary/HentaiAtHome_1.2.5/data
2015-04-21T07:50:39Z [info] 2. Delete the file hath.db
2015-04-21T07:50:39Z [info] 3. Restart the client.
2015-04-21T07:50:39Z [info] The system should now rebuild the database.
2015-04-21T07:50:39Z [info] ***************************************************************************************************************
2015-04-21T07:50:39Z [info]
2015-04-21T07:50:39Z [ERROR] {org.hath.base.HentaiAtHomeClient.dieWithError(HentaiAtHomeClient.java:284)} Critical Error: Failed to load the database.
2015-04-21T07:50:39Z [info] Shutting down...
2015-04-21T07:50:39Z [info] No hard feelings
2015-04-21T07:50:39Z [info] Logging ended.

edit: Thinking about running the executable on an old ssd... over usb 2.0. My gut says this is a bad idea but the quicker reads might make up for the slower transfer output.

This post has been edited by chise87: Apr 21 2015, 13:54

Fuck, I forgot Debian uses dash as its main shell. That thing must run under bash, just type bash to enter a bash shell and tun it.

Fuck, I forgot Debian uses dash as its main shell. That thing must run under bash, just type bash to enter a bash shell and tun it.

On both commands,

diff: /dev/fd/63: No such file or directory
diff: /dev/fd/62: No such file or directory

Is your machine a general purpose computer? That seems like an embedded system.

Anyhow, changed to not use any clever bash extensions (this can run under dash, i've tested):

CODE

find /mnt -type d -perm -111 >/tmp/`id -u`perm111.left
find /mnt -type d -perm /111 >/tmp/`id -u`perm111.right
find /mnt -type d -perm -444 >/tmp/`id -u`perm444.left
find /mnt -type d -perm /444 >/tmp/`id -u`perm444.right
diff /tmp/`id -u`perm111*
diff /tmp/`id -u`perm444*
rm /tmp/`id -u`perm*

Is your machine a general purpose computer? That seems like an embedded system.

Anyhow, changed to not use any clever bash extensions (this can run under dash, i've tested):

CODE

find /mnt -type d -perm -111 >/tmp/`id -u`perm111.left
find /mnt -type d -perm /111 >/tmp/`id -u`perm111.right
find /mnt -type d -perm -444 >/tmp/`id -u`perm444.left
find /mnt -type d -perm /444 >/tmp/`id -u`perm444.right
diff /tmp/`id -u`perm111*
diff /tmp/`id -u`perm444*
rm /tmp/`id -u`perm*

It's a nas. Specially the ix2-dl.

diff /tmp/`id -u`perm111* gives

1829a1830
> /mnt/pools/A/A0/lost+found
1844a1846,1847
> /mnt/system/cups
> /mnt/system/cups/tmp
1868a1872
> /mnt/system/opt/apps/.activation
1964a1969
> /mnt/system/lost+found
1988a1994
> /mnt/etc/lost+found
2001a2008
> /mnt/etc/cups/ssl
2007a2015,2016
> /mnt/etc/lvm/archive
> /mnt/etc/lvm/backup
2015a2025
> /mnt/apps/etc/cups/ssl
2070a2081
> /mnt/apps/etc/ssl/private
2235a2247,2248
> /mnt/apps/mnt/system/cups
> /mnt/apps/mnt/system/cups/tmp
2250a2264
> /mnt/apps/run/lvm
2254a2269
> /mnt/apps/run/lock/lvm
8558a8574
> /mnt/boot/lost+found

diff /tmp/`id -u`perm444* gives

1829a1830
> /mnt/pools/A/A0/lost+found
1844a1846,1847
> /mnt/system/cups
> /mnt/system/cups/tmp
1965a1969
> /mnt/system/lost+found
1989a1994
> /mnt/etc/lost+found
2002a2008
> /mnt/etc/cups/ssl
2008a2015,2016
> /mnt/etc/lvm/archive
> /mnt/etc/lvm/backup
2016a2025
> /mnt/apps/etc/cups/ssl
2071a2081
> /mnt/apps/etc/ssl/private
2236a2247,2248
> /mnt/apps/mnt/system/cups
> /mnt/apps/mnt/system/cups/tmp
2251a2264
> /mnt/apps/run/lvm
2255a2269
> /mnt/apps/run/lock/lvm
8559a8574
> /mnt/boot/lost+found

edit: Added my old ssd. New location is /ssd/HentaiAtHome_1.2.5

This post has been edited by chise87: Apr 22 2015, 01:32

You haven't updated java recently, have you? Especially if it is java 8, java 8 still has pretty poor performance (read: it's buggy as hell) on non-standard systems.

Anyhow, check if fsck can find any problems in the filesystem:

umount the disk at /mnt/pools/A/A0/

then run fsck on it, e.g. fsck /dev/sdb (do not copy and paste it from here, run it on the disk that was mounted on that place)

and see if fsck outputs any errors. If fsck find problematic inodes it will add move them to that lost+found directory.

You haven't updated java recently, have you? Especially if it is java 8, java 8 still has pretty poor performance (read: it's buggy as hell) on non-standard systems.

Anyhow, check if fsck can find any problems in the filesystem:

umount the disk at /mnt/pools/A/A0/

then run fsck on it, e.g. fsck /dev/sdb (do not copy and paste it from here, run it on the disk that was mounted on that place)

and see if fsck outputs any errors. If fsck find problematic inodes it will add move them to that lost+found directory.

I just recently installed java on the nas with whatever recent version for armv5 processors was on the oracle site.

java -version
java version "1.7.0_75"

Also fsck output.

root@ix2-dl:/# fsck /dev/sdb1
fsck 1.42.5 (29-Jul-2012)
e2fsck 1.42.5 (29-Jul-2012)
/dev/sdb1: clean, 7543/1957888 files, 965596/7816406 blocks

This post has been edited by chise87: Apr 22 2015, 08:23

I did run H@H on an arm v4 with openjdk 7, which is pretty close to what you have.

Try mounting the disk again chown'ing all files to the user hath (chown -R hath /mnt/pools/A/A0/), and try again.

If it does not (probably won't as we did not change anything) work run h@H as follows:

CODE

strace java -jar HentaiAtHome.jar 2>&1 | grep open

It will print all system calls to open a file and present the return value of the call. We checked for most filesystem problems so we are safe that the filesystem will give us meaningful answers to system calls.

(if you do not have strace, install it)

Goddamnit. I give up.

root@ix2-dl:/# apt-get install strace
E: Could not open lock file /var/lib/dpkg/lock - open (2: No such file or directory)
E: Unable to lock the administration directory (/var/lib/dpkg/), are you root?

Can't installl sudo because same reason above.
apt-get update doesn't work because same reason above.
Googling tells me that I have another package manager running and halting that process might corrupt something. I don't want to fuck up my iscsi setup.

Goddamnit. I give up.
Can't installl sudo because same reason above.
apt-get update doesn't work because same reason above.
Googling tells me that I have another package manager running and halting that process might corrupt something. I don't want to fuck up my iscsi setup.

Maybe you have an update-script running in the background. You could also check if some processes currently access this file. Using the command fuser /var/lib/dpkg/lock returns the pid of the process.