[Luka]

I think this may point to the problem. The VM may not have enough RAM allocated to it.

How big are the sequence files that you are trying to work with BTW?

I was suspecting that, but I managed to process big sequence files so far using other programs, using different files I have done QC, trimming, assembly , alignment.
Now I just wanted to process some files, thus I created a test file containing 4 150bp sequences, that should be doable.

However, there must be some limitation in this remote connection, which is causing this error.

Can you run the command "free -g" (without quotes of course) while logged into your VM and send the output? That should tell us how much RAM is allocated to your VM.

total used free shared buffers cached
Mem: 3 2 0 0 0 1
-/+ buffers/cache: 1 2
Swap: 5 2 3

This does not look good?

3) Luka: Even without a sysadmin you should be able to install your own copy of fastX tools (from the web page I posted earlier). It may be frustrating but, heck, welcome to the "wonderful" world of bioinformatics.

I guess that, but I does not seem easy, however , it has it charms. Anyway, I truly appreciate the effort you've put into this.
My goal is actually to have a computer with Unix so we can install and use these tools directly on it, rather then like this.

[GenoMax]

Your VM has 3 GB of RAM allocated to it which is probably ok for QC type things (which you have already done). Since you have already done alignments that amount of RAM has not been an issue (perhaps you are working with a small genome).

Is there someone else you can ask to verify the fastx_toolkit install through their account?

[Luka]

Your VM has 3 GB of RAM allocated to it which is probably ok for QC type things (which you have already done). Since you have already done alignments that amount of RAM has not been an issue (perhaps you are working with a small genome).

Is there someone else you can ask to verify the fastx_toolkit install through their account?

Hi, thank you for your post.

As suggested, I actually talked to IT people responsible for the installation of the fastax_toolkit and their reply was that this error has nothing to do with my remote access and that "buffer overflow detected" means that there's a bug in the program that causes it to crash.

And I have some more info I guess it can be useful, it is a RHEV (KVM) virtual machine with 4 GB RAM and 60 GB disk. It is running stock Fedora 17 with FASTX 0.0.13 from the official Fedora package repository (fastx_toolkit-0.0.13-6.fc17.x86_64).

[GenoMax]

Sounds like if you want to use the fastx_toolkit you have to take the extra bit of effort and install it in your home directory as Rick had suggested. You can even get the precompiled binary for 64-bit Linux so all you would need to do is un-zip the archive.

 
wget http://hannonlab.cshl.edu/fastx_toolkit/fastx_toolkit_0.0.13_binaries_Linux_2.6_amd64.tar.bz2

tar -xvjf fastx_toolkit_0.0.13_binaries_Linux_2.6_amd64.tar.bz2

When that works (no obvious reason why it should not) then you can go back to your IT and suggest that they use the original package to redo the install.

[Luka]

Sounds like if you want to use the fastx_toolkit you have to take the extra bit of effort and install it in your home directory as Rick had suggested. You can even get the precompiled binary for 64-bit Linux so all you would need to do is un-zip the archive.

 
wget http://hannonlab.cshl.edu/fastx_toolkit/fastx_toolkit_0.0.13_binaries_Linux_2.6_amd64.tar.bz2

tar -xvjf fastx_toolkit_0.0.13_binaries_Linux_2.6_amd64.tar.bz2

When that works (no obvious reason why it should not) then you can go back to your IT and suggest that they use the original package to redo the install.

Hi, before doing what suggested I forwarded your suggestion to my IT and their reply was simply that to reinstall the toolkit to my home directory will not solve this issue.

[westerman]

Hi, before doing what suggested I forwarded your suggestion to my IT and their reply was simply that to reinstall the toolkit to my home directory will not solve this issue.

So what is their solution? FastX does not have an obvious bug. The rest of us use it all of the time without problems. Certainly I can take the 1-sequence file you gave us and run fastx_renamer on it without problem. If you can not do the same then the problem is clearly in your or your IT's department setup.