doc: add a short thing on kernel client troubleshooting.

I just noticed this sitting uncommitted in my tree.

Signed-off-by: Greg Farnum <gregory.farnum@dreamhost.com>

diff --git a/doc/dev/kernel-client-troubleshooting.rst b/doc/dev/kernel-client-troubleshooting.rst
new file mode 100644 (file)
index 0000000..03ba067
--- /dev/null
+++ b/doc/dev/kernel-client-troubleshooting.rst
@@ -0,0 +1,17 @@
+============
+ Kernel client troubleshooting (FS)
+============
+
+If there is an issue with the cephfs kernel client, the most important thing is
+figuring out whether the problem is with the client or the MDS. Generally,
+this is easy to work out. If the kernel client broke directly, there
+will be output in dmesg. Collect it and any appropriate kernel state. If
+the problem is with the MDS, there will be hung requests that the client
+is waiting on. Look in /sys/kernel/debug/ceph/*/ and cat the mdsc file to
+get a listing of requests in progress. If one of them remains there, the
+MDS has probably "forgotten" it.
+We can get hints about what's going on by dumping the MDS cache:
+ceph mds tell 0 dumpcache /tmp/dump.txt
+
+And if high logging levels are set on the MDS, that will almost certainly
+hold the information we need to diagnose and solve the issue.