Open GoogleCodeExporter opened 8 years ago
What steps will reproduce the problem?
From php connect to memcachedb via Memcache() class
1. add a key with a value
2. try to add again the same key
>> Have you modified your key and value size in memcachedb.h?
What is the expected output? What do you see instead?
-normaly this should return false
Result: PHP Notice: add(): marked server 'localhost:21211' as failed in ...
>> Can you just telnet it to make sure the daemon is available?
What version of the product are you using? On what operating system?
Berkeley DB 4.6.18
libevent 1.4.0-beta
memcachedb-0.1.0
Linux
>> That's exactly what we use.
Original comment by stv...@gmail.com
on 15 Jan 2008 at 2:13
in memcachedb.h
#define END_LEN 5001
#define END_LEN_STR "5000"
-ps aux
root 6514 0.0 12.2 115656 64148 ? Ssl Jan15 1:01 ./memcachedb -p
21211 -d -r -u root -H ./env -N
php test:
<?php
$memcache = new Memcache;
$memcache->addServer('localhost', 21211);
$memcache->add('keya','vala');
echo $memcache->get('keya');
$memcache->add('keya','valb');
result:
Result: valaPHP Notice: add(): marked server 'localhost:21211' as failed in
...
?>
Original comment by cocorult...@gmail.com
on 16 Jan 2008 at 9:24
This might be relevant to the issue above,
Telneting into memcachedb 0.1.0:
root@x:~# telnet localhost 21211
Trying 127.0.0.1...
Connected to localhost.
Escape character is '^]'.
add a1 0 0 5
aaaaa
STORED
add a1 0 0 6
bbbbbb
NOT STORED
Telneting into memcached 1.2.2
root@x:~# telnet localhost 11211
Trying 127.0.0.1...
Connected to localhost.
Escape character is '^]'.
add a1 0 0 5
aaaaa
STORED
add a1 0 0 6
bbbbbb
NOT_STORED
There's a notable difference between the response from the servers on the
second add:
memcached replies with "NOT_STORED"
memcachedb replies with "NOT STORED" (notice the missing underscore)
this probably confuses the client since it's not the standard memcached
response:
http://code.sixapart.com/svn/memcached/trunk/server/doc/protocol.txt
Original comment by razvan....@gmail.com
on 21 Jan 2008 at 9:47
This is bug, and we now have a new bugfix release.
Original comment by stv...@gmail.com
on 21 Jan 2008 at 10:38
Thank you for the extraordinarily quick response.
Keep up the great work!
Original comment by razvan....@gmail.com
on 21 Jan 2008 at 12:51
Tested new release, bug resolved, thanks!
This issue can be closed.
Original comment by cocorult...@gmail.com
on 21 Jan 2008 at 1:12
Hi. I *think* that this code fix still needs to be checked in.
I was using 1.2.1-beta and found this bug myself. I then checked out the code
this
morning from http://memcachedb.googlecode.com/svn/trunk/ and found that line
663 of
memcachedb.c still has NOT_STORED.
Is there a step between the releases that cocorultimid got and subversion that
has
been missed?
Original comment by enda.far...@gmail.com
on 4 Feb 2009 at 2:34
Original issue reported on code.google.com by
cocorult...@gmail.com
on 14 Jan 2008 at 1:44