-
Notifications
You must be signed in to change notification settings - Fork 189
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
inconsistent referencing of $TMPDIR ? #287
Comments
Hi Jason, Can you manually set $TMPDIR to a subdirectory of your current working On 25 October 2016 at 10:19, Jason Kwong notifications@github.com wrote:
|
The weird thing is this didn't use to fail. But it did since I upgraded to Roary 3.7.0 two days ago. Does The issue is that our logins are managed by ActiveDirectory and staff are in the default domain and students are in the STUDENT domain, so their The simplest solution is for me to change my setting of |
Any idea what version you were on previously? Or when you might have last updated? It will help narrow things down, because there have been only very small changes over the last few months, and none of them related to temp directories. Theres no version of File::Temp set in CPAN.One thing to note though is that File::Temp is part of core Perl since 5.20.2, before that it would have come from CPAN, so just double check nobody has done an update to Perl? |
Hello Andrew. Was this File::Temp issue ever resolved? I may be having the same problem though I'm on CentOS. My roary version is the latest ("roary -w" says 1.007001). I always get the error, "BLAST Database error: No alias or index file found for protein database". Looking for the cause, I see that the blast database does not exist and it was makeblastdb that failed upstream. I logged the blast related commands and it looks like each command is using a different temp directory name. I see valid, non-empty protein fasta files getting created in one temp directory, but not the temp directory where makeblastdb runs. And, I don't know if this is cause or effect, but I see that my _clustered file is empty, and that is the file that roary passes to "makeblastdb -in". Thanks for your help. |
Hi Jason,
Torsten has found that linuxbrew has a non functional version of blast
2.6.0 with makeblastdb segment faulting. The solution is to install blast
through a different route.
Andrew
…On 16 Mar 2017 18:07, "JasonRafeMiller" ***@***.***> wrote:
Hello Andrew. Was this File::Temp issue ever resolved? I may be having the
same problem though I'm on CentOS. My roary version is the latest ("roary
-w" says 1.007001). I always get the error, "BLAST Database error: No alias
or index file found for protein database". Looking for the cause, I see
that the blast database does not exist and it was makeblastdb that failed
upstream. I logged the blast related commands and it looks like each
command is using a different temp directory name. I see valid, non-empty
protein fasta files getting created in one temp directory, but not the temp
directory where makeblastdb runs. And, I don't know if this is cause or
effect, but I see that my _clustered file is empty, and that is the file
that roary passes to "makeblastdb -in". Thanks for your help.
—
You are receiving this because you commented.
Reply to this email directly, view it on GitHub
<#287 (comment)>,
or mute the thread
<https://github.com/notifications/unsubscribe-auth/AABeV3xe4jToNpctoT7B_Ipl79uWpij7ks5rmXpogaJpZM4KfvtC>
.
|
@andrewjpage I have changed our login scripts to avoid our |
Hi Andrew,
Thanks very much for your work on
Roary
- it's fantastic and we use it all the time.Just recently (in the last week), I've been getting an error:
I wonder if it is related to inconsistent referencing of
$TMPDIR
?In the output above:
Unfortunately, the way that our user accounts are set up, our usernames have a
\
in the middle which is quite annoying.The text was updated successfully, but these errors were encountered: