liu4lin / spymemcached

Automatically exported from code.google.com/p/spymemcached
0 stars 0 forks source link

Binary connection factory will not recover from a failed connection #189

Open GoogleCodeExporter opened 8 years ago

GoogleCodeExporter commented 8 years ago
What version of the product are you using? On what operating system?
spymemcached 2.7, jmemcached 1.0.0, Mac OS X

If a memcached crashes, the default (text) connection will recover after 
failing a single operation.  The binary connection will not recover, and will 
continue to throw RuntimeExceptions.

I've attached a test case that runs three different tests, all with 
FailureMode.Retry.  First, it sets a key.  Then it "power cycles" the 
memcached.  Then it attempts to set and get a key.

testCrashedCacheDefaultStrict -- ASCII factory, no failures allowed (fail)
testCrashedCacheDefaultLoose -- ASCII factory, one failure allowed (pass)
testCrashedCacheBinary -- binary factory, one failure allowed (fail)

I would expect all three to pass.  I have explicitly told spymemcached that I 
am okay with retries, and no operations were in flight at the time of the 
"power cycle"

Original issue reported on code.google.com by stevensc...@gmail.com on 22 Jul 2011 at 1:03

Attachments:

GoogleCodeExporter commented 8 years ago
Note that this is likely related to Issue #108 and Issue #181, however I have 
chosen to create a new issue because this one has a self-contained test case 
that only involves one server (rather than an entire cluster, like the other 
issues) and so may be clearer to work in isolation.

Thank you for taking a look at this!

Original comment by stevensc...@gmail.com on 22 Jul 2011 at 1:04

GoogleCodeExporter commented 8 years ago
Oops, there is a small typo in that the binary test actually uses 
DefaultConnectionFactory rather than BinaryConnectionFactory.

Original comment by stevensc...@gmail.com on 22 Jul 2011 at 1:14

GoogleCodeExporter commented 8 years ago
Thanks for the test!  I hope this has actually already been fixed with recent 
updates, but we'll find out.

Original comment by ingen...@gmail.com on 16 Oct 2011 at 2:13

GoogleCodeExporter commented 8 years ago
This still seems to be a problem :-(

Original comment by stevensc...@gmail.com on 4 Mar 2013 at 7:42

GoogleCodeExporter commented 8 years ago
Which version did you re-observe it with @stevenschlansker?

Original comment by ingen...@gmail.com on 4 Mar 2013 at 8:42

GoogleCodeExporter commented 8 years ago
My test still does not pass against master of dustin/java-memcached-client

Original comment by stevensc...@gmail.com on 4 Mar 2013 at 8:46

GoogleCodeExporter commented 8 years ago
Hi Steven,

so you are running 2.8.11?

Original comment by michael....@gmail.com on 5 Mar 2013 at 6:18