Kodi Community Forum
v18 library updates - partial by source / remote / skip none reachable device - Printable Version

+- Kodi Community Forum (https://forum.kodi.tv)
+-- Forum: Support (https://forum.kodi.tv/forumdisplay.php?fid=33)
+--- Forum: General Support (https://forum.kodi.tv/forumdisplay.php?fid=111)
+---- Forum: OS independent / Other (https://forum.kodi.tv/forumdisplay.php?fid=228)
+---- Thread: v18 library updates - partial by source / remote / skip none reachable device (/showthread.php?tid=368692)



library updates - partial by source / remote / skip none reachable device - mstep - 2022-06-22

Hi,

I have 2 devices each defined as their own video sources in Kodi 18.
However, one device is not running 24/7, so when I start a library update, it takes forever because it waits for the timeout per file. In the debug log this is 3.1 seconds per file. Since the library is larger this takes a long time.
Why doesn't Kodi notice that the IP of the device is unreachable and skips all files to this source/device without a new request?

I am looking for a simple way to update only the source of the active device directly.

You can do a remotely library update see here https://kodi.wiki/view/HOW-TO:Remotely_update_library .
But this is only a library update over all sources. Can this be done via the API also for source A or source B?

Any other suggestions?


RE: library updates - partial by source / remote / skip none reachable device - Klojum - 2022-06-22

(2022-06-22, 18:06)mstep Wrote: Why doesn't Kodi notice that the IP of the device is unreachable and skips all files to this source/device without a new request?

It's not about IP addresses, but it's about the file protocol you use.

You can try to add the following to your advancedsettings.xml file:
xml:
<samba>
<clienttimeout>5</clienttimeout> <!-- timeout (in seconds) -->
</samba>
or
xml:
<network>
<curlclienttimeout>10</curlclienttimeout> <!-- Timeout in seconds for libcurl (http/ftp) connections -->
<nfstimeout>5</nfstimeout> <!-- Added in v19- Timeout in seconds for NFS access, 0 = wait forever (previous behaviour) default is 5s -->
</network>