[Tickets #5874] Re: Mnemo2 locks up when encountering large notes

bugs at bugs.horde.org bugs at bugs.horde.org
Fri Nov 16 20:19:20 UTC 2007


DO NOT REPLY TO THIS MESSAGE. THIS EMAIL ADDRESS IS NOT MONITORED.

Ticket URL: http://bugs.horde.org/ticket/?id=5874
-----------------------------------------------------------------------
 Ticket             | 5874
 Updated By         | jwinius at umrk.to
 Summary            | Mnemo2 locks up when encountering  large notes
 Queue              | Mnemo
 Version            | 2.1.1
 Type               | Bug
 State              | Feedback
 Priority           | 2. Medium
 Owners             | 
-----------------------------------------------------------------------


jwinius at umrk.to (2007-11-16 12:19) wrote:

After some more experimentation I've made an important discovery. Although
the problem does have to do with the size of the notes, it now appears not
to be a problem with the size of the individual notes. Rather, it seems to
be caused by the total size of all of the notes.

Not wanting to wait for a solution, I wrote a perl script to process my
horde2 database dump and print out the sizes of the individual mnemo notes.
My intention was to identify the notes that were too large and chop them up
into smaller ones so that I could get mnemo2 to work again. However, the
printout showed that a number of other notes, that were being successfully
read by mnemo2, were much larger than the 34.6K size limit I mentioned
earlier -- they were up to 77K in size.

Another test showed that if I started with a known non-working situation
-- leaving that larger 34.6K note intact on the end -- and then deleted a
few smaller notes (instead of making the 34.6K note smaller), mnemo2 would
suddenly start to work again.

I calculated the total number of bytes and found that if all of the notes
amounted to 1,252,799 bytes or less, mnemo2 would work. However, if the
total was 1,266,774 bytes or more, mnemo2 would not work. The actual limit
is somewhere in between.

I'm guessing, but could it be that this bug has something to do with the
mnemo2 preview function?



More information about the bugs mailing list