Serverphorums.com - Memcached

The Very Versatile Thomas Sabo Necklace (no replies)

nini01 — Mon, 11 Mar 2013 06:36:09 +0100

The
thomas sabo necklace and other jewellery lines are aimed at both women and men. This is part of the large success of this high quality silver jewellery. There is a little something for everybody no matter what their taste. Part of being successful in the jewellery business is being able to cater to a very wide audience.

One of the most endearingly popular products is the charm necklace. You can choose a necklace or a bracelet to adorn with multiple decorative charms, and there are plenty out there to choose from. These
thomas sabo uk come from every interest and aspect of life that you can think of. Everyone can have a necklace that reflects their individual taste and experiences. In today's society, we are into being unique and showing others who we are. This is the great thing about the Thomas Sabo Necklace and charm line.

All necklaces and bracelets are sterling silver of the highest quality. Depending on the length of the Thomas Sabo Necklace, it has many different weights. You can start at around dhs80 in price and work up to dhs250. After you choose a necklace, there are silver connectors that you choose from based again on weight.

If you buy a
thomas sabo charms necklace, it is advised to wear no more than three charms. Because of the weight of the silver, your neck can get weighed down pretty quick. It isn't easy limiting yourself. You can get as many charms as you want with a bracelet. With a Thomas Sabo Necklace, you can create a unique collection of charms and then just alternate. This will give you a unique look everyday or whenever you feel like you want to change. You can also have the feeling that you are wearing new jewellery every day. You will feel different when being able to have a fresh look whenever you want it.

Today, many men are going for the
thomas sabo sale Necklace. There are many masculine charms and charms that carry a more chunky character. There is a line of Gothic charms that appeal to a lot of men as well. There is no reason in today's day and age, why jewellery lines can't cater more to the individual sides of men. Thomas Sabo has done that with his ever growing line of sterling silver jewellery and necklaces.

Anyone can accessorise with their Thomas Sabo Necklace. You can buy rings, earrings, key chains, bracelets, and the endless number of charms. The statement to be made with this jewellery is almost infinitely variable.

If you are looking for Thomas Sabo necklaces, many retailers on the internet sell this fine line of sterling silver jewellery. Many suppliers offer everything from the wide collection of necklaces and charms, to bracelets, earrings, rings, and more. You can find whatever you are looking for when it comes to Thomas Sabo. Get in on a high quality silver jewellery line that can be as unique as you are. There really is something special for everyone.

Making a purchase could not be easier. Just browse our store, and add any items that you wish to buy into the shopping cart.We sell a range of jewellery, including Thomas Sabo necklace and other brands. Please visit our website at thomas sabo charms uk.

per-key cache misses (no replies)

Peter Phaal — Thu, 28 Oct 2010 18:40:07 +0200

Hi All,

The following article describes how to use sFlow instrumentation in
Memcached to identify the top keys responsible for cache misses:
http://blog.sflow.com/2010/10/memcached-missed-keys.html

For anyone interested in trying out sFlow in Memcached, the patch can
be downloaded from github:
http://github.com/sflow/memcached

Cheers,
Peter

curious questions (1 reply)

Minh Doan — Wed, 27 Oct 2010 01:10:13 +0200

Hi folks,

Sorry for the following stupid question.

I've got the idea of memcached. Memcached data are stored in multiple
memcached servers which run independently. Whenever you need to lookup a
(key/value), just calculate a hash to find the server id, and then call a
lookup-service on the server. So:

what are the advantages of memcached against a normal server which store
data in ram-memory?
Thanks
---
Minh

Memcached underutilized (3 replies)

amix — Sat, 23 Oct 2010 14:30:05 +0200

We have recently added a new memcached server, but it's underutilized
or at least not as much utilized as our other memcached servers. The
difference between this new server and the older server are following:
* memcached 1.4.5 is run (instead of memcached 1.2.2)
* Ubuntu 10 is run (instead of Ubuntu 8)

Has something drastically changed between 1.2.2 and 1.4.5? Or has
something changed between Ubuntu 8 and Ubuntu 10? Anybody else having
a similar issue?

Here is usage patterns from these two servers. Each of them have 3GB
available, but the server running 1.4.5 only uses around 300MB. We are
unsure what the cause is. These servers are run like this:

memcached -m 3072 -k -p 11221 -u plurk -l 192.168.0.35 -c 10240

Regards,
amix

[ ( '192.168.0.35:11221 (1)',
{ 'accepting_conns': '1',
'auth_cmds': '0',
'auth_errors': '0',
'bytes': '71301667',
'bytes_read': '68596588527',
'bytes_written': '215641263049',
'cas_badval': '0',
'cas_hits': '0',
'cas_misses': '0',
'cmd_flush': '0',
'cmd_get': '246655883',
'cmd_set': '62909113',
'conn_yields': '0',
'connection_structures': '2526',
'curr_connections': '2262',
'curr_items': '132819',
'decr_hits': '0',
'decr_misses': '0',
'delete_hits': '2297822',
'delete_misses': '23286159',
'evictions': '0',
'get_hits': '163185291',
'get_misses': '83470592',
'incr_hits': '25466',
'incr_misses': '0',
'limit_maxbytes': '3221225472',
'listen_disabled_num': '0',
'pid': '7319',
'pointer_size': '64',
'reclaimed': '36224211',
'rusage_system': '8341.840000',
'rusage_user': '3173.390000',
'threads': '4',
'time': '1287781046',
'total_connections': '139794',
'total_items': '62899221',
'uptime': '514507',
'version': '1.4.5'}),
( '192.168.0.70:11221 (1)',
{ 'bytes': '2879886510',
'bytes_read': '65548028841',
'bytes_written': '179395226207',
'cmd_get': '277646387',
'cmd_set': '68425776',
'connection_structures': '2553',
'curr_connections': '2291',
'curr_items': '3907837',
'evictions': '0',
'get_hits': '193391910',
'get_misses': '84254477',
'limit_maxbytes': '3221225472',
'pid': '17580',
'pointer_size': '64',
'rusage_system': '6113.240000',
'rusage_user': '3507.560000',
'threads': '1',
'time': '1287781045',
'total_connections': '138271',
'total_items': '68416527',
'uptime': '514517',
'version': '1.2.2'})]

Matrix in memcache (4 replies)

Sreejith S — Fri, 29 Oct 2010 03:10:06 +0200

Hi all..

I have dense matrix (large sized matrix).Because of its size i could not be
able to kept in memory.So i would like to use memcache.I configured
memcached and is working.
But how can i represent this Matrix input as a key,value pair.Is there any
need to use any Db like MongoDb...

Pls help me to find out the solution

Thank you
Sreejith

memcached out of the box versus behavior tweaks (1 reply)

Patrick Galbraith — Thu, 28 Oct 2010 03:30:07 +0200

Hi all,

Just wondering about a few things while pondering - what are most people
using in terms of memcached out of the box - starting it up with a
particular size, but not setting any particular behavior, or what
settings have people modified in the behavior of the client that they've
found useful for their installations?

Also, for large installations, what have people found works better - one
large instance or smaller multiple instances? I'm thinking if you were
to use buckets, that the latter wouldn't be as necessary.

Thoughts?

cannot assign requested address (99) (2 replies)

Ilan Berkner — Fri, 22 Oct 2010 08:50:17 +0200

We have a very busy high traffic site and recently started using memached to
store frequently accessed queries. We put memcached on each php application
server so that the connection is local to the system.

Today, during one of our peek times, we started getting this error and had
to disable the caching of queries:

[20-Oct-2010 12:05:57] PHP Warning: Memcache::connect() [ href='memcache.connect'>memcache.connect]: Can't connect to
127.0.0.1:11211, Cannot assign requested address (99)

In searching online for this error, I came across several articles, but
nothing that I though was relevant to our case. I tried increasing the
number of connections from 1024 (default) to 5000 but the error continued.

Does anyone have any suggestions / thoughts of where to look?

stats information is below:

STAT pid 13425
STAT uptime 1940
STAT time 1287591991
STAT version 1.4.5
STAT pointer_size 64
STAT rusage_user 6.443020
STAT rusage_system 25.109182
STAT curr_connections 15
STAT total_connections 340578
STAT connection_structures 118
STAT cmd_get 357224
STAT cmd_set 136334
STAT cmd_flush 0
STAT get_hits 326956
STAT get_misses 30268
STAT delete_misses 0
STAT delete_hits 0
STAT incr_misses 0
STAT incr_hits 0
STAT decr_misses 0
STAT decr_hits 0
STAT cas_misses 0
STAT cas_hits 0
STAT cas_badval 0
STAT auth_cmds 0
STAT auth_errors 0
STAT bytes_read 21641880
STAT bytes_written 41425437
STAT limit_maxbytes 1073741824
STAT accepting_conns 1
STAT listen_disabled_num 0
STAT threads 4
STAT conn_yields 0
STAT bytes 2925203
STAT curr_items 19070
STAT total_items 136334
STAT evictions 0
STAT reclaimed 12
END

Most efficient way of modifying value (5 replies)

Jorge — Tue, 26 Oct 2010 23:30:07 +0200

Hello!

I would like to use memcached as a fast cache for a hospital order
database.

Each order has a location (5 possible values), procedure (about 100
possible values),
and patient status (3 different values).

So, my thinking was for each order, create a key as follows:

(location code)_(procedure code)_(status code)

and the value would be a list of all orders matching a given code.

So, doing a query on a given location, procedure and status will just
entail reading the value from the key.

My question is: what is the best way of modifying the value, as an in-
memory list.

Can this be done on the memcached server, or do I need the client to
pull the value out,
modify it, and put it back in?

Thanks!

next planned release of memcached (2 replies)

Pavel Kushnirchuk — Tue, 19 Oct 2010 18:40:11 +0200

Folks,

May be anybody know, when will be a next planned release of memcached.
Especially I am interested in a release of Win32 branch.

Enhancement of memcached (2 replies)

Pavel Kushnirchuk — Mon, 18 Oct 2010 23:50:05 +0200

Sorry may be my question is a bit stupid.
I need to extend a memcached with very simple additional
functionality(allow to listen memcached on multiple IP addresses).
Could I do all necessary changes in Win32 branch? Will all my changes
to go to any of next releases?

PHP and memcached issues (1 reply)

christophorosp — Mon, 18 Oct 2010 17:40:26 +0200

Hi all,
Congratulations about your wonderful community

I'm kindly asking your help to resolve an issue:

Im struggling to use memcache with PHP on my server.
I run Red Hat Enterprise Linux Server release 5.5 with php-memcache
3.0.4 and memcached 1.4.5 installed.

The PHP implementation i have setup returns an exception that it
cannot store data to memcached
I don't have access to the code since its encrypted but i would like
to verify that my configuration is correct.
Is anyone willing to help me? Would be much appreciated

Thank you
Chris

memcached server for listening of list of IP addresses (no replies)

Pavel Kushnirchuk — Mon, 18 Oct 2010 17:40:26 +0200

Is it planned for any of future releases of memcached to include
support for listening of list of IP addresses.

how do you find out which version of memcached you have installed? (1 reply)

Ilan Berkner — Mon, 18 Oct 2010 16:50:35 +0200

Running stats command but its not showing listen_disabled_num? (no replies)

Ilan Berkner — Mon, 18 Oct 2010 16:40:20 +0200

Memory capacity overflow of memcached (3 replies)

yashushi — Wed, 20 Oct 2010 09:30:13 +0200

Hi,

My team try to use memcached to provide PHP caching recently but we
find out there is a memory overflow problem.

Here is the steps we have tested.
(1)Create memcached with the max. memory is 4MB .
(2)Add 900KB data from web to memcached 3 times.
(3)Add 700KB data from web to memcached.
(4)Add 500KB data from web to memcached.
(5)Add 300KB data from web to memcached.
(6)The new data still can transfer to memcached when new data size is
greater than memcached remaining space size.

We use "stats" to check "bytes" and it shows the memory size is over
4MB.
Could you help to clarify what is going on?
Thank you.

Regards,
Yashushi

xmemcached 1.2.6.1 released and update the java memcached client benchmark (2 replies)

dennis — Mon, 18 Oct 2010 11:10:09 +0200

hi all

xmemcached is an open-source memcached client for java,i am pleased
to announce the release of Xmemcached version 1.2.6.1,it is a minor
version fixed bugs.if you use maven ,you cloud use it as

com.googlecode.xmemcached
xmemcached
1.2.6.1

And also i have updated the java memcached client performance
benchmark ,you can check it on

http://xmemcached.googlecode.com/svn/trunk/benchmark/benchmark.html

homepage: http://code.google.com/p/xmemcached/
download: http://code.google.com/p/xmemcached/downloads/list
wiki: http://code.google.com/p/xmemcached/wiki/User_Guide

Any suggestions are welcome,thanks.

Updated: Problems about two memcached java clients: spy and gwhalin (10 replies)

Shi Yu — Sun, 17 Oct 2010 10:00:01 +0200

Follow up my previous question. I tried to read those 6 million
records out. Both API are capable, but the spymemcache API
is faster (16 minutes) than the Whalin's v2.5.1 distribution (24
minutes) (http://github.com/gwhalin/Memcached-Java-Client/downloads)
..Since the spymemcache cannot insert that much, I came to a strange
hybrid settings using Whalin's API to store, and using spymemcache to
read ...

I am really curious why spy cannot store up to 6 million...

On Sat, Oct 16, 2010 at 12:59 PM, Shi Yu wrote:
> Hi,
>
> I have two problems when using memcached java clients the spymemcached
> (http://code.google.com/p/spymemcached/) and the gwhalin java client
> for memcached (http://github.com/gwhalin/Memcached-Java-Client). I
> found that the spymemcached failed to store more than 4.3 million
> records sometimes 3.7 million (please see my code below). There was no
> error no exception, but simply the code automatically stopped at the
> 4.3 million and didn't even hit the final line. In contrast, the
> gwhalin java client was able to insert 6 million records without
> problem, however, comparing the speed of inserting the first 4 million
> records that the gwhalin client is much slower than the spymemcached.
> The memcached server is set up using the following command
> "./memcached -d -m 4000 127.0.0.1 -p 11211" and I think there is no
> problem at the server side. What is the problem here, should I adjust
> any settings? Thanks?
>
> -Shi
>
>
> //spymemcached code
> public static void main(String[] args) throws Exception {
> MemcachedClient mc=new MemcachedClient(new
> InetSocketAddress("ocuic32.research", 11211));
> mc.flush();
> System.out.println("Memchaced flushed ...");
> int count = 0;
> for(int i=0;i<6000000;i++){
> String a = "String"+i;
> String b = "Value"+i;
>
> mc.add(a,i,(String) b);
> count ++;
> if (String.valueOf(count).endsWith("00000"))
> System.out.println(count+ " elements added.");
> }
>
> System.out.println("done "+ count +" records inserted");
> //spymemcached aint able to get this line
> }
>
>
>
> //gwhalin memcached code
> public static void main(String[] args) throws Exception {
> BasicConfigurator.configure();
> String[] servers = { "ocuic32.research:11211" };
> SockIOPool pool = SockIOPool.getInstance();
> pool.setServers( servers );
> pool.setFailover( true );
> pool.setInitConn( 10 );
> pool.setMinConn( 5 );
> pool.setMaxConn( 250 );
> pool.setMaintSleep( 30 );
> pool.setNagle( false );
> pool.setSocketTO( 3000 );
> pool.setAliveCheck( true );
> pool.initialize();
>
> MemcachedClient mcc = new MemcachedClient();
> mcc.flushAll();
> int count = 0;
> int maxlength = 0;
> //while((line=br.readLine())!=null){
>
> for(int i=0;i<6000000;i++){
> String a = "String"+i;
> String b = "Value"+i;
> String sha1_ad1 = AeSimpleSHA1.SHA1(a);
> mcc.set(sha1_ad1,(String) b);
> count ++;
> if (String.valueOf(count).endsWith("00000"))
> System.out.println(count+ " elements added.");
>
> }
>
> System.out.println("done "+ count +" records
> inserted"); //gwhalin 's client is able to get this line, but very
> slow
> }
>

Problems about two memcached java clients: spy and gwhalin (7 replies)

Shi Yu — Sun, 17 Oct 2010 09:30:01 +0200

Hi,

I have two problems when using memcached java clients the spymemcached
(http://code.google.com/p/spymemcached/) and the gwhalin java client
for memcached (http://github.com/gwhalin/Memcached-Java-Client). I
found that the spymemcached failed to store more than 4.3 million
records sometimes 3.7 million (please see my code below). There was no
error no exception, but simply the code automatically stopped at the
4.3 million and didn't even hit the final line. In contrast, the
gwhalin java client was able to insert 6 million records without
problem, however, comparing the speed of inserting the first 4 million
records that the gwhalin client is much slower than the spymemcached.
The memcached server is set up using the following command
"./memcached -d -m 4000 127.0.0.1 -p 11211" and I think there is no
problem at the server side. What is the problem here, should I adjust
any settings? Thanks?

-Shi

//spymemcached code
public static void main(String[] args) throws Exception {
MemcachedClient mc=new MemcachedClient(new
InetSocketAddress("ocuic32.research", 11211));
mc.flush();
System.out.println("Memchaced flushed ...");
int count = 0;
for(int i=0;i<6000000;i++){
String a = "String"+i;
String b = "Value"+i;

mc.add(a,i,(String) b);
count ++;
if (String.valueOf(count).endsWith("00000"))
System.out.println(count+ " elements added.");
}

System.out.println("done "+ count +" records inserted");
//spymemcached aint able to get this line
}

//gwhalin memcached code
public static void main(String[] args) throws Exception {
BasicConfigurator.configure();
String[] servers = { "ocuic32.research:11211" };
SockIOPool pool = SockIOPool.getInstance();
pool.setServers( servers );
pool.setFailover( true );
pool.setInitConn( 10 );
pool.setMinConn( 5 );
pool.setMaxConn( 250 );
pool.setMaintSleep( 30 );
pool.setNagle( false );
pool.setSocketTO( 3000 );
pool.setAliveCheck( true );
pool.initialize();

MemcachedClient mcc = new MemcachedClient();
mcc.flushAll();
int count = 0;
int maxlength = 0;
//while((line=br.readLine())!=null){

for(int i=0;i<6000000;i++){
String a = "String"+i;
String b = "Value"+i;
String sha1_ad1 = AeSimpleSHA1.SHA1(a);
mcc.set(sha1_ad1,(String) b);
count ++;
if (String.valueOf(count).endsWith("00000"))
System.out.println(count+ " elements added.");

}

System.out.println("done "+ count +" records
inserted"); //gwhalin 's client is able to get this line, but very
slow
}

Key longer than 250 characters (3 replies)

Shi Yu — Sat, 16 Oct 2010 05:50:02 +0200

Hi

I wonder whether the key in Memcached could be longer than 250
characters. Is there any way to configure this? I feel the 250
characters is too small because imagine if there is a dna sequence
data as a key, it could exceed that limit.

Shi

Deleting cache objects by group (1 reply)

Abhilash — Fri, 15 Oct 2010 18:20:10 +0200

Hey,

Is there a way I can delete the cache objects by group?
I am using net.spy.memcached.MemcachedClient as my Java client. But it
offers a functionality to delete cache by key.

I am looking for something where I can group my keys and delete the
group whenever I need to delete all of them.

Thanks
Abhilash

Silicon Valley and Memcache (no replies)

Dave Haverstick — Fri, 15 Oct 2010 01:50:06 +0200

We are recruiting for two Data Warehousing/Cloud Computing software
companies in the South Bay of Silicon Valley. One is a block from the
Cal-train.

They are looking for Sr Software Engineers with a C++ and Memcached
background and Sr QA Engineers.

Please call me or suggest and we can discuss the specifics.

Thank you

Dave Haverstick
Senior Technical Recruiter

Office: 650-763-8758 x200
Email: daveh@triadgroup.com
LinkedIn: www.linkedin.com/in/davetriadgroup

1MB limit on object size (3 replies)

Sakuntala — Tue, 19 Oct 2010 20:40:16 +0200

Hi,

I am using
memcached server version 1.4.4 on Linux and
spy memcached client version 2.4.2

I performed a test by storing and retrieving objects of size 1.8MB and
3.6MB. I am able to retrieve the objects successfully without any
error. My load tests show some timeout exceptions when memcached
server is under heavy load, which is expected.

Memcached wiki website states that one of the reasons we should not be
using memcached is if the object size exceeds 1MB. I want to know if
this is a hard limit, or just that memcached is not efficient in that
scenario due to the storage model.

The reason for my question is that I have just one object(which is a
list) which might exceed 1MB in future, but currently, it is not
exceeding the limit. I considered storing it in several chunks and
using bulkGet() to retrieve the multiple small objects. The response
times are not efficient in case of bulkGet(). I should not be using
memcached for this object( and should look into other caching
mechanisms) but dont want to change the implementation till next
release. Please suggest.

Thanks,

Issue 160 in memcached: problems when save objects into the memcached (no replies)

Anonymous User — Thu, 14 Oct 2010 12:00:11 +0200

Status: New
Owner: ----
Labels: Type-Defect Priority-Medium

New issue 160 by jameshu1984: problems when save objects into the memcached
http://code.google.com/p/memcached/issues/detail?id=160

now i'm working on a project with memcached in php, and now i have a
problem stuck me for some days

when i tried to save object into the memcached, the value stored in the
memcached didnt match the thing i put into

suppose now i want to save a object

class Test_Obj {

public $a = 1;
public $b = 2;
public $c = 3;

/*
*/
public function __sleep(){
print 'test';
return array('a', 'b');
}
/*
*/
public function __unset($name){
unset($this->$name);
}
}

$testObj = new Test_Obj();
$memcached = new Memcached();
$memcachedKey = 'test_obj';
$memcached->setOption(Mist_Memcached::OPT_SERIALIZER,Memcached::SERIALIZER_PHP);
$memcached->set($memcachedKey, $testObj, 10);
$data = $memcached->get($memcachedKey);
var_dump($data);

i think the output should be like:
testobject(Test_Obj)#252 (3) { ["a"]=> int(2) ["b"]=> int(2)}

but actually the result is different:
testobject(Test_Obj)#252 (3) { ["a"]=> int(2) ["b"]=> int(2) ["c"]=>
int(3) }

what i couldnt understand is, it does invoke the __sleep function, print
the 'test' string, but the value stored in the memcached still have the 'c'
property.

and i tested for the second time.

i added some unset code like:
$testObj = new Test_Obj();
$memcached = new Memcached();
$memcachedKey = 'test_obj';
$memcached->setOption(Mist_Memcached::OPT_SERIALIZER,Memcached::SERIALIZER_PHP);
unset($testObj->c);
$memcached->set($memcachedKey, $testObj, 10);
$data = $memcached->get($memcachedKey);
var_dump($data);

the output is the same as above.
testobject(Test_Obj)#252 (3) { ["a"]=> int(2) ["b"]=> int(2) ["c"]=>
int(3) }

i cant understand why i unset the property 'c' but still get the the
memcached value with the property 'c'

is there anyone can help me? honestly thanks

Flush a particular Cache Client (no replies)

circuitsdorf — Wed, 13 Oct 2010 22:20:04 +0200

I know that the flush_all command will invalidate the contents of the
entire memcached instance.
I am using a Java-based Spy Memcached client, which has CacheClient
names configured.
Suppose I wanted to flush a specific CacheClient?
How would I do that?

Is memcache add() atomic on a multithreaded memcached? (17 replies)

elSchrom — Sun, 17 Oct 2010 18:10:07 +0200

Hi everyone,

we have the following situation: due to massive simultaneous inserts
in mysql on possibly identical primary keys, we use the atomic
memcache add() as a semaphore. In a few cases we observed the
behaviour, that two simultaneous add() using the same key from
different clients both returned true (due to consistent hashing the
key has to be on the same machine).

Is it now possible, that the multithreaded memcached does return true
on two concurrent add() on the same key, if the requests are handled
by two different threads on the same machine?

Any information on this would be appreciated.

Kind regard,

Jerome

Monitoring Memcached w/ Evident ClearStone (no replies)

Tim Sneed — Wed, 13 Oct 2010 18:10:08 +0200

To Fellow Memcached Users,

Evident Software would like to invite you to the following webinar.
Date/Time: October 21, 2010 (Thursday) – 2pm EST

Duration: 45 minutes

Webinar Topic
Attendees will learn how Evident ClearStone can be used to monitor and
manage your Cassandra, Memcached, or Oracle Coherence application
environments.

Who should attend?
Anyone who is responsible for developing, testing, or supporting
applications that use NoSQL products like Cassandra or distributed
caching technologies like Oracle Coherence, Memcached.

Sign-up
Go to http://www.evidentsoftware.com/category/news-events/events and
sign up today.
Inform your colleagues about this event.

Website: www.evidentsoftware.com
Email: info@evidentsoftware.com
Twitter: @evidentsoftware

Proxy Manager ? (1 reply)

Dominic Webb — Wed, 13 Oct 2010 16:30:05 +0200

As I am readin and using memcached more (by the way amazing software)
I am looking at sharding, scaling and the whole consistent hashing
stuff. Our environment is PHP 5.3+ by the way.

Im wondering if anything exists that can act as a manager for what is
stored and on which server. Specifically is there not some sort of
proxy server that handles the request and decides which server to
goto?

I appreciate this puts a layer of latency infront of memcached but
(hypothetically if something doesn't exist) if this is a bad idead
then why?

Perhaps a more theoretical post than anything else so thanks in
advance.

Dom W

Gigs of swap being used, but why? (5 replies)

Gavin Hamill — Sat, 09 Oct 2010 19:30:05 +0200

Hi :)

We have a small farm of 8 memcached boxes (Dell R210s, Debian Lenny
x64) with 8GB RAM. I've just noticed that a) there is swap configured
on these machines (probably my bad) and b) many gigabytes of swap is
being used.

memcache09:~# ps faux
USER PID %CPU %MEM VSZ RSS TTY STAT START TIME
COMMAND
[kernel processes omitted]
root 1 0.0 0.0 10312 296 ? Ss Feb18 1:19 init
[2]
root 1013 0.0 0.0 16920 236 ? S --daemon
root 1943 0.0 0.0 126172 1532 ? Sl Feb18 0:23 /usr/
sbin/rsyslogd -c3
root 1954 0.0 0.0 3796 228 ? Ss Feb18 0:00 /usr/
sbin/acpid
root 1970 0.0 0.0 48864 516 ? Ss Feb18 0:00 /usr/
sbin/sshd
root 10929 0.0 0.0 66068 3112 ? Ss 13:10 0:00 \_
sshd: root@pts/0
root 10931 0.0 0.0 18756 1836 pts/0 Ss 13:10 0:00
\_ -bash
root 10962 0.0 0.0 15980 1052 pts/0 R+ 13:27
0:00 \_ ps faxu
root 2011 0.0 0.0 19804 372 ? Ss Feb18 0:09 /usr/
sbin/cron
root 2028 0.0 0.0 3796 208 tty1 Ss+ Feb18 0:00 /sbin/
getty 38400 tty1
root 2029 0.0 0.0 3796 208 tty2 Ss+ Feb18 0:00 /sbin/
getty 38400 tty2
root 2030 0.0 0.0 3796 208 tty3 Ss+ Feb18 0:00 /sbin/
getty 38400 tty3
root 2032 0.0 0.0 3796 208 tty4 Ss+ Feb18 0:00 /sbin/
getty 38400 tty4
root 2034 0.0 0.0 3796 208 tty5 Ss+ Feb18 0:00 /sbin/
getty 38400 tty5
root 2036 0.0 0.0 3796 208 tty6 Ss+ Feb18 0:00 /sbin/
getty 38400 tty6
nobody 31164 1.0 37.4 3521308 3072284 ? Sl Apr28 2413:10 /usr/
bin/memcached -m 2928 -p 11212 -u nobody -c 8192
nobody 31179 0.0 1.7 429080 144436 ? Ssl Apr28 50:44 /usr/
bin/memcached -m 192 -p 11213 -u nobody -c 8192
nobody 31381 0.2 54.6 5427948 4478588 ? Sl Apr28 594:38 /usr/
bin/memcached -m 4680 -p 11211 -u nobody -c 8192

memcache09:~# free
total used free shared buffers
cached
Mem: 8199492 8099508 99984 0 16052
16900
-/+ buffers/cache: 8066556 132936
Swap: 9936160 3620640 6315520

The 'swappiness' proc knob was on its default of 60, but I just don't
get it... the VIRT columns of all running process don't add up to
anywhere near (physical RAM + swap in use) (they add to under 10GB)

It can't even be disk cache because there's only 660MB in use on the
rootfs...

Filesystem 1K-blocks Used Available Use% Mounted on
/dev/sda1 230528596 669032 218149388 1% /

I fully expect if I restart the 3 memcache processes (not something I
want to do since it will hurt live performance) the swap will zero,
but that doesn't solve the root cause.

Any ideas would be warmly welcomed! :)

Cheers,
Gavin

memcached timeout error because of slow response (6 replies)

dl4ner — Wed, 27 Oct 2010 12:20:08 +0200
Hi,

we expirience memcache slow responses. (rare, only 3..5x per 24h)
I tracked the "mecache timeout errors" down to the following:

36278 13:52:46.020698 192.168.43.129 -> 192.168.43.133 MEMCACHE get
key
36299 13:52:46.061557 192.168.43.133 -> 192.168.43.129 TCP memcache >
47127 [ACK] Seq=1 Ack=75 Win=68 Len=0 TSV=623862695
40947 13:52:47.024100 192.168.43.129 -> 192.168.43.133 TCP 47127 >
memcache [FIN, ACK] Seq=75 Ack=1 Win=501 Len=0 TSV=324
41125 13:52:47.061603 192.168.43.133 -> 192.168.43.129 TCP memcache >
47127 [ACK] Seq=1 Ack=76 Win=68 Len=0 TSV=623862945
45256 13:52:48.316037 192.168.43.133 -> 192.168.43.129 MEMCACHE VALUE
key
limited during capture]
45257 13:52:48.316096 192.168.43.129 -> 192.168.43.133 TCP 47127 >
memcache [RST] Seq=76 Win=0 Len=0
45262 13:52:48.316055 192.168.43.133 -> 192.168.43.129 MEMCACHE
MEMCACHE Continuation[Packet size limited during capture]
45263 13:52:48.316134 192.168.43.129 -> 192.168.43.133 TCP 47127 >
memcache [RST] Seq=76 Win=0 Len=0

so it is clear: the memcache accepts the "get key", but does not
respond during the timeout value of 1 second.
and no, we do not have packet loss here, becaues this is the trace of
the same time on the memcached server:

34602 13:52:46.021765 192.168.43.129 -> 192.168.43.133 MEMCACHE get
key
34606 13:52:46.062537 192.168.43.133 -> 192.168.43.129 TCP memcache >
47127 [ACK] Seq=1 Ack=75 Win=68 Len=0 TSV=
38633 13:52:47.025136 192.168.43.129 -> 192.168.43.133 TCP 47127 >
memcache [FIN, ACK] Seq=75 Ack=1 Win=501 Len=
38745 13:52:47.062530 192.168.43.133 -> 192.168.43.129 TCP memcache >
47127 [ACK] Seq=1 Ack=76 Win=68 Len=0 TSV=
40769 13:52:48.316909 192.168.43.133 -> 192.168.43.129 MEMCACHE VALUE
key
limited during capture]
40772 13:52:48.316919 192.168.43.133 -> 192.168.43.129 MEMCACHE
MEMCACHE Continuation[Packet size limited during
40785 13:52:48.317124 192.168.43.129 -> 192.168.43.133 TCP 47127 >
memcache [RST] Seq=76 Win=0 Len=0
40788 13:52:48.317179 192.168.43.129 -> 192.168.43.133 TCP 47127 >
memcache [RST] Seq=76 Win=0 Len=0

now some data:
memcached-server is running 2.6.31-gentoo-r4, memcache-1.4.5 and a
mysql instance.
it is a single quadcore@2.5 GHz w/ 8 GB RAM, about 6 GB free.
it is configured for 2GB of RAM, and uses about 1300 pages. no
evictions occure.
during the time when those sporadically slow responses (I had a trace
of a 20 second-response for a set command) a lot of other memcache
requests in the meantime are processed normally.
the network is a gbit switch, both servers use serverboard with
onboard gbit nics (intel),
and the traffic during that is quite low (about 20Mbit/s and 3000
packets/s on the memcached-server).
also load and io are low.

any hints ?

regards.

Werner.

using memcached for php session management - high load site (4 replies)

iberkner — Wed, 13 Oct 2010 02:50:02 +0200
Hi All,

Newbie question, looked around, not clear on the answers.

4 php application servers, currently using file system for session
management

switched over to single 2GB memcached server for dealing with php
sessions

high load site, 3mm page views a day

our cpu and server load was getting "out of control" when I switched
over to the memcached for php session management

how many concurrent connections can memcached handle? how do i check
to see if in fact this is the bottleneck?

any other information regarding what settings / stat values to look
for would be helpful.

I'm considering using a memcaced server for each php app server but
then it wouldn't be a centralized situation.