Significant performance degradation on a server with big RAM #5352

PhantomYdn · 2015-11-20T02:32:40Z

Guys,

We got a server with 128Gb RAM and 32 CPUs. But OrientDB is running even slower than on 1 CPU/1Gb account on digital ocean. The mostly it's observable during db startup: it's doing something for a couple of minutes (vs. few minutes on small environment). According to thread dump it's freezing here:

"main" prio=10 tid=0x00007f4e2c00a000 nid=0x18a3 runnable [0x00007f4e33a93000]
   java.lang.Thread.State: RUNNABLE
    at java.io.RandomAccessFile.write0(Native Method)
    at java.io.RandomAccessFile.write(RandomAccessFile.java:500)
    at java.io.RandomAccessFile.writeLong(RandomAccessFile.java:1071)
    at com.orientechnologies.orient.core.storage.impl.local.paginated.wal.ODiskWriteAheadLog.writeMasterRecord(ODiskWriteAheadLog.java:1216)
    at com.orientechnologies.orient.core.storage.impl.local.paginated.wal.ODiskWriteAheadLog.log(ODiskWriteAheadLog.java:884)
    at com.orientechnologies.orient.core.storage.impl.local.paginated.wal.OAbstractWriteAheadLog.logFullCheckpointStart(OAbstractWriteAheadLog.java:67)
    at com.orientechnologies.orient.core.storage.impl.local.OAbstractPaginatedStorage.makeFullCheckpoint(OAbstractPaginatedStorage.java:1631)
    at com.orientechnologies.orient.core.storage.impl.local.OAbstractPaginatedStorage.synch(OAbstractPaginatedStorage.java:1205)
    at com.orientechnologies.orient.core.metadata.schema.OSchemaShared$1.call(OSchemaShared.java:1217)
    at com.orientechnologies.orient.core.db.OScenarioThreadLocal.executeAsDistributed(OScenarioThreadLocal.java:67)
    at com.orientechnologies.orient.core.metadata.schema.OSchemaShared.saveInternal(OSchemaShared.java:1210)
    at com.orientechnologies.orient.core.metadata.schema.OSchemaShared.releaseSchemaWriteLock(OSchemaShared.java:640)
    at com.orientechnologies.orient.core.metadata.schema.OClassImpl.releaseSchemaWriteLock(OClassImpl.java:1808)
    at com.orientechnologies.orient.core.metadata.schema.OClassImpl.releaseSchemaWriteLock(OClassImpl.java:1803)
    at com.orientechnologies.orient.core.metadata.schema.OPropertyImpl.releaseSchemaWriteLock(OPropertyImpl.java:1109)
    at com.orientechnologies.orient.core.metadata.schema.OPropertyImpl.setCustom(OPropertyImpl.java:757)
    at com.orientechnologies.orient.core.metadata.schema.OPropertyImpl.setCustom(OPropertyImpl.java:68)

It's interesting, but JVM instance limited just be few Gb.

Any ideas?

PhantomYdn · 2015-11-20T03:57:50Z

Forgot to mention timings also for db creaton: it depends but commonly it also took >1 min to create just empty database.
OrientDB version 2.1.5

andrii0lomakin · 2015-11-20T16:15:37Z

Hi,
It is funny. I will explain reason.
Now we need to force sync data for each class creation because you have huge amount of memory it will take a lot to make of time to force sync whole cache. Is it possible to create classes before head to avoid such behaviour ?

PhantomYdn · 2015-11-20T16:26:55Z

Andrey,

Not sure that I understood your explanation... But let me mention that runtime working with schema (creation/deletion of classes and properties) is common use case for Orienteer.

Aaa.... Just checked performance of classes/properties creation: it takes about 3 sec in avg. That's crazy...

andrii0lomakin · 2015-11-20T16:39:10Z

I think I have solution for you.
I will set max limit of memory for write cache it should solve your problems.
Meantime could you limit disk cache size ?

PhantomYdn · 2015-11-20T17:48:11Z

Andrey,
Sounds good. But it looks like short-term solution. Am I right?
And can we please discuss reasons of that in more details by skype?
I can't beleive that we have to limit OrientDB to make it faster:) And even more: as I said we limited OrientDB just by 2gb of RAM (by -Xms2g -Xmx2g) and empty db takes less then 10Mb on hard-drive: so problem looks really strange from 3rd side.

andrii0lomakin · 2015-11-20T17:57:28Z

Sure, let's do it on Monday.

On Fri, Nov 20, 2015 at 7:48 PM Ilia Naryzhny notifications@github.com
wrote:

Andrey,
Sounds good. But it looks like short-term solution. Am I right?
And can we please discuss reasons of that in more details by skype?
I can't beleive that we have to limit OrientDB to make it faster:) And
even more: as I said we limited OrientDB just by 2gb of RAM (by -Xms2g
-Xmx2g) and empty db takes less then 10Mb on hard-drive: so problem looks
really strange from 3rd side.

—
Reply to this email directly or view it on GitHub
#5352 (comment)
.

Best regards,
Andrey Lomakin, R&D lead.
OrientDB Ltd

twitter:@Andrey_Lomakin linkedin:https://ua.linkedin.com/in/andreylomakin

lvca · 2015-12-04T09:38:24Z

Any feedback on this?

andrii0lomakin · 2015-12-04T10:59:51Z

We agreed with Iya to work on this right after 2.2

On Fri, Dec 4, 2015 at 11:38 AM Luca Garulli notifications@github.com
wrote:

Any feedback on this?

—
Reply to this email directly or view it on GitHub
#5352 (comment)
.

Best regards,
Andrey Lomakin, R&D lead.
OrientDB Ltd

twitter:@Andrey_Lomakin linkedin:https://ua.linkedin.com/in/andreylomakin

andrii0lomakin · 2015-12-25T18:45:22Z

HI,
Could you try now from build in develop branch ?

PhantomYdn · 2015-12-25T21:06:37Z

Should I make some configuration changes? According to diff there are some config properties...

andrii0lomakin · 2015-12-26T10:00:18Z

Hi Ilia,
No you do not have to change default settings.

On Fri, Dec 25, 2015 at 11:06 PM Ilia Naryzhny notifications@github.com
wrote:

Should I make some configuration changes? According to diff there are some
config properties...

—
Reply to this email directly or view it on GitHub
#5352 (comment)
.

Best regards,
Andrey Lomakin, R&D lead.
OrientDB Ltd

twitter:@Andrey_Lomakin linkedin:https://ua.linkedin.com/in/andreylomakin

andrii0lomakin · 2016-01-18T10:04:02Z

Ilya, any updates on this issue ?

PhantomYdn · 2016-01-18T10:08:34Z

Not yet - sorry. Pretty hard to "bring" SNAPSHOT version (which is not compatible with latest Orienteer for now) to production.
Any due dates for that? When are you going to have some RC for 2.2?

PhantomYdn · 2016-03-22T16:06:31Z

Still no ability to test on prod for 2.2-SNAPSHOT.

But there is significant degradation of this issue in 2.1.13 (probably introduced somewhere here: 2.1.10-2.1.13). If before it took ~20 sec for schema update - now it can take up to 2 minutes(!!!) for simple update like adding new class.

andrii0lomakin · 2016-03-22T16:20:18Z

@PhantomYdn could you profile such update and send us yourkit snapshot if possible ? I do not think that issues are related , so would be good to send to me for example by email.

PhantomYdn · 2016-03-22T17:23:21Z

During class creation it hanging here:

"qtp81628611-111" #111 prio=5 os_prio=0 tid=0x00007fc8cffc2800 nid=0x61ad runnable [0x00007fca2a1c3000]
   java.lang.Thread.State: RUNNABLE
        at com.orientechnologies.orient.core.metadata.schema.OSchemaShared.acquireSchemaReadLock(OSchemaShared.java:618)
        at com.orientechnologies.orient.core.metadata.schema.OClassImpl.acquireSchemaReadLock(OClassImpl.java:1866)
        at com.orientechnologies.orient.core.metadata.schema.OPropertyImpl.acquireSchemaReadLock(OPropertyImpl.java:1093)
        at com.orientechnologies.orient.core.metadata.schema.OPropertyImpl.getLinkedType(OPropertyImpl.java:398)
        at com.orientechnologies.orient.core.metadata.schema.OImmutableProperty.<init>(OImmutableProperty.java:60)
        at com.orientechnologies.orient.core.metadata.schema.OImmutableClass.<init>(OImmutableClass.java:86)
        at com.orientechnologies.orient.core.metadata.schema.OImmutableSchema.<init>(OImmutableSchema.java:39)
        at com.orientechnologies.orient.core.metadata.schema.OSchemaShared.saveInternal(OSchemaShared.java:1228)
        at com.orientechnologies.orient.core.metadata.schema.OSchemaShared.releaseSchemaWriteLock(OSchemaShared.java:642)
        at com.orientechnologies.orient.core.metadata.schema.OSchemaShared.releaseSchemaWriteLock(OSchemaShared.java:631)
        at com.orientechnologies.orient.core.metadata.schema.OSchemaShared.doCreateClass(OSchemaShared.java:1050)
        at com.orientechnologies.orient.core.metadata.schema.OSchemaShared.createClass(OSchemaShared.java:415)
        at com.orientechnologies.orient.core.metadata.schema.OSchemaShared.createClass(OSchemaShared.java:400)
        at com.orientechnologies.orient.core.metadata.schema.OSchemaShared.createClass(OSchemaShared.java:275)
        at com.orientechnologies.orient.core.metadata.schema.OSchemaProxy.createClass(OSchemaProxy.java:71)

And then a catched:

"qtp81628611-117" #117 prio=5 os_prio=0 tid=0x00007fc928004800 nid=0x6351 runnable [0x00007fca29378000]
   java.lang.Thread.State: RUNNABLE
        at java.io.RandomAccessFile.write0(Native Method)
        at java.io.RandomAccessFile.write(RandomAccessFile.java:489)
        at java.io.RandomAccessFile.writeInt(RandomAccessFile.java:1031)
        at com.orientechnologies.orient.core.storage.impl.local.paginated.wal.ODiskWriteAheadLog.writeMasterRecord(ODiskWriteAheadLog.java:1210)
        at com.orientechnologies.orient.core.storage.impl.local.paginated.wal.ODiskWriteAheadLog.log(ODiskWriteAheadLog.java:884)
        at com.orientechnologies.orient.core.storage.impl.local.paginated.wal.OAbstractWriteAheadLog.logFullCheckpointStart(OAbstractWriteAheadLog.java:67)
        at com.orientechnologies.orient.core.storage.impl.local.OAbstractPaginatedStorage.makeFullCheckpoint(OAbstractPaginatedStorage.java:1738)
        at com.orientechnologies.orient.core.storage.impl.local.OAbstractPaginatedStorage.synch(OAbstractPaginatedStorage.java:1288)
        at com.orientechnologies.orient.core.metadata.schema.OSchemaShared$1.call(OSchemaShared.java:1219)
        at com.orientechnologies.orient.core.db.OScenarioThreadLocal.executeAsDistributed(OScenarioThreadLocal.java:71)
        at com.orientechnologies.orient.core.metadata.schema.OSchemaShared.saveInternal(OSchemaShared.java:1212)
        at com.orientechnologies.orient.core.metadata.schema.OSchemaShared.releaseSchemaWriteLock(OSchemaShared.java:642)
        at com.orientechnologies.orient.core.metadata.schema.OSchemaShared.releaseSchemaWriteLock(OSchemaShared.java:631)
        at com.orientechnologies.orient.core.metadata.schema.OSchemaShared.doCreateClass(OSchemaShared.java:1050)
        at com.orientechnologies.orient.core.metadata.schema.OSchemaShared.createClass(OSchemaShared.java:415)

During property creation:

"qtp81628611-113" #113 prio=5 os_prio=0 tid=0x00007fc8ef9f5000 nid=0x626d runnable [0x00007fca2a1c3000]
   java.lang.Thread.State: RUNNABLE
        at java.io.RandomAccessFile.write0(Native Method)
        at java.io.RandomAccessFile.write(RandomAccessFile.java:489)
        at java.io.RandomAccessFile.writeLong(RandomAccessFile.java:1049)
        at com.orientechnologies.orient.core.storage.impl.local.paginated.wal.ODiskWriteAheadLog.writeMasterRecord(ODiskWriteAheadLog.java:1212)
        at com.orientechnologies.orient.core.storage.impl.local.paginated.wal.ODiskWriteAheadLog.log(ODiskWriteAheadLog.java:884)
        at com.orientechnologies.orient.core.storage.impl.local.paginated.wal.OAbstractWriteAheadLog.logFullCheckpointStart(OAbstractWriteAheadLog.java:67)
        at com.orientechnologies.orient.core.storage.impl.local.OAbstractPaginatedStorage.makeFullCheckpoint(OAbstractPaginatedStorage.java:1738)
        at com.orientechnologies.orient.core.storage.impl.local.OAbstractPaginatedStorage.synch(OAbstractPaginatedStorage.java:1288)
        at com.orientechnologies.orient.core.metadata.schema.OSchemaShared$1.call(OSchemaShared.java:1219)
        at com.orientechnologies.orient.core.db.OScenarioThreadLocal.executeAsDistributed(OScenarioThreadLocal.java:71)
        at com.orientechnologies.orient.core.metadata.schema.OSchemaShared.saveInternal(OSchemaShared.java:1212)
        at com.orientechnologies.orient.core.metadata.schema.OSchemaShared.releaseSchemaWriteLock(OSchemaShared.java:642)
        at com.orientechnologies.orient.core.metadata.schema.OClassImpl.releaseSchemaWriteLock(OClassImpl.java:1883)
        at com.orientechnologies.orient.core.metadata.schema.OClassImpl.releaseSchemaWriteLock(OClassImpl.java:1878)
        at com.orientechnologies.orient.core.metadata.schema.OClassImpl.addProperty(OClassImpl.java:2387)
        at com.orientechnologies.orient.core.metadata.schema.OClassImpl.createProperty(OClassImpl.java:769)

Looks realy strange... It's unclear: is it just one call for random file access or there multiple invocations. But so drastical degradation was not here before.

andrii0lomakin · 2016-03-23T08:26:25Z

Well it is clear for me, and already fixed in issue which we discuss now
:-)).

On Tue, Mar 22, 2016 at 7:23 PM Ilia Naryzhny notifications@github.com
wrote:

During class creation it hanging here:

"qtp81628611-111" #111 prio=5 os_prio=0 tid=0x00007fc8cffc2800 nid=0x61ad runnable [0x00007fca2a1c3000]
java.lang.Thread.State: RUNNABLE
at com.orientechnologies.orient.core.metadata.schema.OSchemaShared.acquireSchemaReadLock(OSchemaShared.java:618)
at com.orientechnologies.orient.core.metadata.schema.OClassImpl.acquireSchemaReadLock(OClassImpl.java:1866)
at com.orientechnologies.orient.core.metadata.schema.OPropertyImpl.acquireSchemaReadLock(OPropertyImpl.java:1093)
at com.orientechnologies.orient.core.metadata.schema.OPropertyImpl.getLinkedType(OPropertyImpl.java:398)
at com.orientechnologies.orient.core.metadata.schema.OImmutableProperty.(OImmutableProperty.java:60)
at com.orientechnologies.orient.core.metadata.schema.OImmutableClass.(OImmutableClass.java:86)
at com.orientechnologies.orient.core.metadata.schema.OImmutableSchema.(OImmutableSchema.java:39)
at com.orientechnologies.orient.core.metadata.schema.OSchemaShared.saveInternal(OSchemaShared.java:1228)
at com.orientechnologies.orient.core.metadata.schema.OSchemaShared.releaseSchemaWriteLock(OSchemaShared.java:642)
at com.orientechnologies.orient.core.metadata.schema.OSchemaShared.releaseSchemaWriteLock(OSchemaShared.java:631)
at com.orientechnologies.orient.core.metadata.schema.OSchemaShared.doCreateClass(OSchemaShared.java:1050)
at com.orientechnologies.orient.core.metadata.schema.OSchemaShared.createClass(OSchemaShared.java:415)
at com.orientechnologies.orient.core.metadata.schema.OSchemaShared.createClass(OSchemaShared.java:400)
at com.orientechnologies.orient.core.metadata.schema.OSchemaShared.createClass(OSchemaShared.java:275)
at com.orientechnologies.orient.core.metadata.schema.OSchemaProxy.createClass(OSchemaProxy.java:71)

And then a catched:

"qtp81628611-117" #117 prio=5 os_prio=0 tid=0x00007fc928004800 nid=0x6351 runnable [0x00007fca29378000]
java.lang.Thread.State: RUNNABLE
at java.io.RandomAccessFile.write0(Native Method)
at java.io.RandomAccessFile.write(RandomAccessFile.java:489)
at java.io.RandomAccessFile.writeInt(RandomAccessFile.java:1031)
at com.orientechnologies.orient.core.storage.impl.local.paginated.wal.ODiskWriteAheadLog.writeMasterRecord(ODiskWriteAheadLog.java:1210)
at com.orientechnologies.orient.core.storage.impl.local.paginated.wal.ODiskWriteAheadLog.log(ODiskWriteAheadLog.java:884)
at com.orientechnologies.orient.core.storage.impl.local.paginated.wal.OAbstractWriteAheadLog.logFullCheckpointStart(OAbstractWriteAheadLog.java:67)
at com.orientechnologies.orient.core.storage.impl.local.OAbstractPaginatedStorage.makeFullCheckpoint(OAbstractPaginatedStorage.java:1738)
at com.orientechnologies.orient.core.storage.impl.local.OAbstractPaginatedStorage.synch(OAbstractPaginatedStorage.java:1288)
at com.orientechnologies.orient.core.metadata.schema.OSchemaShared$1.call(OSchemaShared.java:1219)
at com.orientechnologies.orient.core.db.OScenarioThreadLocal.executeAsDistributed(OScenarioThreadLocal.java:71)
at com.orientechnologies.orient.core.metadata.schema.OSchemaShared.saveInternal(OSchemaShared.java:1212)
at com.orientechnologies.orient.core.metadata.schema.OSchemaShared.releaseSchemaWriteLock(OSchemaShared.java:642)
at com.orientechnologies.orient.core.metadata.schema.OSchemaShared.releaseSchemaWriteLock(OSchemaShared.java:631)
at com.orientechnologies.orient.core.metadata.schema.OSchemaShared.doCreateClass(OSchemaShared.java:1050)
at com.orientechnologies.orient.core.metadata.schema.OSchemaShared.createClass(OSchemaShared.java:415)

During property creation:

"qtp81628611-113" #113 prio=5 os_prio=0 tid=0x00007fc8ef9f5000 nid=0x626d runnable [0x00007fca2a1c3000]
java.lang.Thread.State: RUNNABLE
at java.io.RandomAccessFile.write0(Native Method)
at java.io.RandomAccessFile.write(RandomAccessFile.java:489)
at java.io.RandomAccessFile.writeLong(RandomAccessFile.java:1049)
at com.orientechnologies.orient.core.storage.impl.local.paginated.wal.ODiskWriteAheadLog.writeMasterRecord(ODiskWriteAheadLog.java:1212)
at com.orientechnologies.orient.core.storage.impl.local.paginated.wal.ODiskWriteAheadLog.log(ODiskWriteAheadLog.java:884)
at com.orientechnologies.orient.core.storage.impl.local.paginated.wal.OAbstractWriteAheadLog.logFullCheckpointStart(OAbstractWriteAheadLog.java:67)
at com.orientechnologies.orient.core.storage.impl.local.OAbstractPaginatedStorage.makeFullCheckpoint(OAbstractPaginatedStorage.java:1738)
at com.orientechnologies.orient.core.storage.impl.local.OAbstractPaginatedStorage.synch(OAbstractPaginatedStorage.java:1288)
at com.orientechnologies.orient.core.metadata.schema.OSchemaShared$1.call(OSchemaShared.java:1219)
at com.orientechnologies.orient.core.db.OScenarioThreadLocal.executeAsDistributed(OScenarioThreadLocal.java:71)
at com.orientechnologies.orient.core.metadata.schema.OSchemaShared.saveInternal(OSchemaShared.java:1212)
at com.orientechnologies.orient.core.metadata.schema.OSchemaShared.releaseSchemaWriteLock(OSchemaShared.java:642)
at com.orientechnologies.orient.core.metadata.schema.OClassImpl.releaseSchemaWriteLock(OClassImpl.java:1883)
at com.orientechnologies.orient.core.metadata.schema.OClassImpl.releaseSchemaWriteLock(OClassImpl.java:1878)
at com.orientechnologies.orient.core.metadata.schema.OClassImpl.addProperty(OClassImpl.java:2387)
at com.orientechnologies.orient.core.metadata.schema.OClassImpl.createProperty(OClassImpl.java:769)

Looks realy strange... It's unclear: is it just one call for random file
access or there multiple invocations. But so drastical degradation was not
here before.

—
You are receiving this because you were assigned.

Reply to this email directly or view it on GitHub
#5352 (comment)

Best regards,
Andrey Lomakin, R&D lead.
OrientDB Ltd

twitter:@Andrey_Lomakin linkedin:https://ua.linkedin.com/in/andreylomakin

PhantomYdn · 2016-03-23T09:02:19Z

That's great:) Thanks! But when we will have some fix? In 2.2 or 2.1.14? And as I said, things became worsen in latest 2.1.X fixes... Any workaround for 2.1?

andrii0lomakin · 2016-03-28T17:13:05Z

Well sorry we do not provide radical changes in design of underlying engine
in hotfixes.
((.

On Wed, Mar 23, 2016 at 11:02 AM Ilia Naryzhny notifications@github.com
wrote:

That's great:) Thanks! But when we will have some fix? In 2.2 or 2.1.14?
And as I said, things became worsen in latest 2.1.X fixes... Any workaround
for 2.1?

—
You are receiving this because you were assigned.

Reply to this email directly or view it on GitHub
#5352 (comment)

Best regards,
Andrey Lomakin, R&D lead.
OrientDB Ltd

twitter:@Andrey_Lomakin linkedin:https://ua.linkedin.com/in/andreylomakin

smolinari · 2016-03-28T17:22:33Z

So the fix is in 2.2, right?

Scott

andrii0lomakin · 2016-03-29T06:04:32Z

Yes, exactly.

andrii0lomakin · 2016-05-06T06:03:11Z

@PhantomYdn Did you have a chance to try it on 2.2 rc1 ?

andrii0lomakin · 2016-05-23T11:10:44Z

Close issue because of no reply. If you reproduce given issue and have information which is requested in issue feel free to reopen it.

PhantomYdn · 2016-05-23T15:40:30Z

Was not able to mobe due to #6157 . Will be able to say something after full transion

lvca added the performance label Nov 20, 2015

lvca assigned andrii0lomakin Nov 20, 2015

lvca added this to the 2.1.x (next hotfix) milestone Nov 20, 2015

lvca modified the milestones: 2.1.x (next hotfix), 2.1.6 Nov 23, 2015

andrii0lomakin added the in progress label Dec 25, 2015

andrii0lomakin modified the milestones: 2.2.0-rc1, 2.1.x (next hotfix) Dec 25, 2015

andrii0lomakin added a commit that referenced this issue Dec 25, 2015

Issue #5352 was fixed but Javadocs and tests need to be added.

19300f0

andrii0lomakin added a commit that referenced this issue Jan 8, 2016

Issue #5352 minor refactoring is performed, Javadoc is added.

00a07a5

andrii0lomakin added the waiting reply label Jan 18, 2016

andrii0lomakin added the resolved label Feb 12, 2016

andrii0lomakin removed the waiting reply label Mar 15, 2016

andrii0lomakin added storage team and removed storage team labels Apr 12, 2016

andrii0lomakin modified the milestones: 2.2.0-rc1, 2.2.0 GA Apr 20, 2016

andrii0lomakin added waiting reply and removed resolved labels May 6, 2016

andrii0lomakin modified the milestones: 2.2.1, 2.2.0 GA May 10, 2016

andrii0lomakin modified the milestones: 2.2.1, 2.2.x (next hotfix) May 23, 2016

andrii0lomakin closed this as completed May 23, 2016

andrii0lomakin removed the waiting reply label May 23, 2016

andrii0lomakin removed this from the 2.2.x (next hotfix) milestone May 23, 2016

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Significant performance degradation on a server with big RAM #5352

Significant performance degradation on a server with big RAM #5352

PhantomYdn commented Nov 20, 2015

PhantomYdn commented Nov 20, 2015

andrii0lomakin commented Nov 20, 2015

PhantomYdn commented Nov 20, 2015

andrii0lomakin commented Nov 20, 2015

PhantomYdn commented Nov 20, 2015

andrii0lomakin commented Nov 20, 2015

lvca commented Dec 4, 2015

andrii0lomakin commented Dec 4, 2015

andrii0lomakin commented Dec 25, 2015

PhantomYdn commented Dec 25, 2015

andrii0lomakin commented Dec 26, 2015

andrii0lomakin commented Jan 18, 2016

PhantomYdn commented Jan 18, 2016

PhantomYdn commented Mar 22, 2016

andrii0lomakin commented Mar 22, 2016

PhantomYdn commented Mar 22, 2016

andrii0lomakin commented Mar 23, 2016

PhantomYdn commented Mar 23, 2016

andrii0lomakin commented Mar 28, 2016

smolinari commented Mar 28, 2016

andrii0lomakin commented Mar 29, 2016

andrii0lomakin commented May 6, 2016

andrii0lomakin commented May 23, 2016

PhantomYdn commented May 23, 2016

Significant performance degradation on a server with big RAM #5352

Significant performance degradation on a server with big RAM #5352

Comments

PhantomYdn commented Nov 20, 2015

PhantomYdn commented Nov 20, 2015

andrii0lomakin commented Nov 20, 2015

PhantomYdn commented Nov 20, 2015

andrii0lomakin commented Nov 20, 2015

PhantomYdn commented Nov 20, 2015

andrii0lomakin commented Nov 20, 2015

lvca commented Dec 4, 2015

andrii0lomakin commented Dec 4, 2015

andrii0lomakin commented Dec 25, 2015

PhantomYdn commented Dec 25, 2015

andrii0lomakin commented Dec 26, 2015

andrii0lomakin commented Jan 18, 2016

PhantomYdn commented Jan 18, 2016

PhantomYdn commented Mar 22, 2016

andrii0lomakin commented Mar 22, 2016

PhantomYdn commented Mar 22, 2016

andrii0lomakin commented Mar 23, 2016

PhantomYdn commented Mar 23, 2016

andrii0lomakin commented Mar 28, 2016

smolinari commented Mar 28, 2016

andrii0lomakin commented Mar 29, 2016

andrii0lomakin commented May 6, 2016

andrii0lomakin commented May 23, 2016

PhantomYdn commented May 23, 2016