Record activity after 703 says 'Your Journal is full"

[douglas_arellanes]

Today I updated to 703, and have tried to use the Journal activity, but I'm getting a message that says 'Your Journal is full' and it won't allow me to take pictures or record video. I went to the journal and removed items one by one, so that it's almost entirely empty.

Any idea of why that might be and how I can fix it?

[redheads]

I have exactly the same problem - moved all journal data to a USB drive and deleted it from the hard drive - still get the message that states "your journal is full" -- what now???

[redheads]

Ok, I used the 'o' key to boot into the last build, and now get a message that 'the disk is full' (despite the fact that I emptied the journal) so I am going to download the offline update (I did the online one last time) and wipe it all and try again, will keep you posted...

[redheads]

Ok, after a totally clean install (backed up all data we wanted to save via usb first!) I have a working machine again. I guess the disk was full with temp files (not in the journal!) and I had no idea how to clean them off!

Hope that helps!

[boomer400]

OK, this occurs also with a 767 upgrade. Has anyone found a solution other than clean install?

[RedEye]

I'm getting this same problem as well after the 767 upgrade. How do I clear out the temp files? Help!

[rwhe]

I seem to have fixed the problem. As far as I can tell, all the "journal full" error means is that your flash drive is getting full. I used the df -h command to determine that my drive had 47M free. I then typed

$ sudo yum clean all

because I had been installing a lot of RPMs. This freed almost another 20M, which seemed to be sufficient, because I'm not getting the error anymore.

Thus, for me, the magic number of free megabytes seems to be between 47M and 66M. Perhaps a nice round 50?

In my experimentation, I also did this:

$ mv .sugar .sugar.old
$ sudo reboot

I still got the error afterwards and I don't recommend this.

[jlvandewalle]

I'm running build 767 and have loaded it a couple of times. I tried the sudo yum clean all command in terminal and it seems to clear some temp file up, but hasn't solved the problem with record, journal is full. It appears that the metadata file is also kinda clogged up, and wonder if I somehow have multipul builds clogging the "hard drive". I've played around with erase commands in terminal, catiously, with no luck. I'd try a clean install, but the download has been problematic. I like the record function and want it back. I even reflashed to build 711? and still no camera. Any heip?

[ektoric]

Would it be possible to post the output of the "df" command?
1. start the Terminal activity
2. df

[jlvandewalle]

Sure can,
mtd0 1048576 902332 87% /
tmpfs 35364 0 0% /dev/shm

And command df -h reveals:

mtd0 1.0G 882M 143M 87% /
tmpfs 35M 0 35 0% / dev/shm

Thanks

[boomer400]

Ektoric,
sorry, I resorted to flashing (2) 767 clean installs from my jumpdrive which gave me something like 47% usage per machine; I just didn't want to mess around too much. In any event, next time I will do a df after some rpms to see what happens and try to post.
It would be useful to have an FAQ/How-To for space/usage issues problem determination.

[ektoric]

Per http://dev.laptop.org/ticket/6758 the latest comment by cjb says that this is expected behaviour. You indeed have less than 200mb free, and I guess Record wants to make sure it has at least 200mb free. Try removing some of the older versions of the OS. I think someone here recently posted excellent suggestions on how to do that.

Edit: Ah yes, it was culseg in viewtopic.php?f=1334&t=162324

[jlvandewalle]

Hey, thanks for the help so far, I follwed the link to culberts tips on dumping files, but I'm afraid I'm missing something that may be intuitive for a more experienced user. Neither the rm or ls commands perform the way I would expect and with the # sign in plce the commands do nothing. I think that the problem is probably that a more experienced user would know that some other info belongs in between slash marks or where the OLD BUILD HASH line is. so, I have a year old g1g1 xo updated online every time I was aware of a new build. I guess I need to get rid of 656, 703, 711. I poked around as far as I;m comfortable sticking these buld numbers into command code lines, but I;m afraid of doing someting stupid. Any futher direction?

[ektoric]

In unix, it is conventional to put a "#" in front of commands when sharing information in forums. This serves two purposes.
1. Usually the prompt for the root user is "#". So this serves as an indicator that "this is a command that you type at the prompt".
2. A leading "#" character also serves as comment in shell scripts. Everything that follows becomes a "comment", and is not executed. This is a small safety net so that if someone cuts and pastes, they don't accidentally execute something they didn't expect or weren't yet ready to execute. For example if part of the command is a placeholder meant to be replaced by something unique to the target, you don't want to cut and paste and run the placeholder string.

As for how to delete old versions, please see this post: How to delete old versions

[jlvandewalle]

Thank you for the quick reply. Your responses are very much appreciated and clear. Please bear with me as I'll probably have more neophyte questions as I work this out. This forum thread has been very useful, even though I haven't resolved the issue, I get the feeling I'm making progress. And learning some stuff.
Thanks again, JL