<br><br><div class="gmail_quote">On Tue, Sep 28, 2010 at 10:07 PM, Simon Andrews <span dir="ltr"><<a href="mailto:simon.andrews@bbsrc.ac.uk">simon.andrews@bbsrc.ac.uk</a>></span> wrote:<br><blockquote class="gmail_quote" style="margin: 0pt 0pt 0pt 0.8ex; border-left: 1px solid rgb(204, 204, 204); padding-left: 1ex;">
I have a fedora 13 box on which I have a remote mounted nfs share over a<br>
fairly slow (10Mb/s) link. I'm then transferring data onto this share<br>
from a different machine using scp.<br>
<br>
The problem is that after scp reports that it's 100% complete the<br>
program will hang for ~20 mins before it will move on to another file.<br>
At this point it can't be killed.<br>
<br>
It looks like the nfs daemon is caching write data (around 2GB of it)<br>
which lets scp think its finished when actually there's loads of data<br>
sitting in a write buffer. The hanging is presumably the time it takes<br>
to flush the buffer (there is a process called nfsiod which is active<br>
during this time and df shows data is still being written).<br>
<br>
Does anyone know how to either make this buffer smaller, or get rid of<br>
it all together so the scp can accruately report on its progress?<br>
<br>
Thanks<br>
<br>
Simon.<br>
<font color="#888888"><br>
--<br>
users mailing list<br>
<a href="mailto:users@lists.fedoraproject.org">users@lists.fedoraproject.org</a><br>
To unsubscribe or change subscription options:<br>
<a href="https://admin.fedoraproject.org/mailman/listinfo/users" target="_blank">https://admin.fedoraproject.org/mailman/listinfo/users</a><br>
Guidelines: <a href="http://fedoraproject.org/wiki/Mailing_list_guidelines" target="_blank">http://fedoraproject.org/wiki/Mailing_list_guidelines</a><br>
</font></blockquote></div><br><a href="http://nfs.sourceforge.net/nfs-howto/ar01s05.html"> This Article</a> contains information on how to adjust the buffer size of NFS and optimise file transfers. Also scp has a -C option to enable compression.<br>