Re: Review Request 59058: SENTRY-1752 HMSFollower gets stuck once it fails to process a notification event

y for HMSFollower to make forward progress is to update notification IDs - otherwise it will process the same event over and over and continue failing. - Alexander Kolbasov On May 8, 2017, 6:07 p.m., Na Li wrote: > > --- &

Re: SQL changes for SENTRY-1726

The major problem with read-modify-write approach is that it is difficult to handle the case with two writers trying to update the value at the same time. If you handle this by adding new rows and having the ID as the primary key, one writer will succeed and another will fail because the key alr

Re: Review Request 58975: SENTRY-1669 HMSFollower should read current processed notification ID from database every time it runs

://reviews.apache.org/r/58975/#comment247368> This is never used - Alexander Kolbasov On May 3, 2017, 9:26 p.m., kalyan kumar kalvagadda wrote: > > --- > This is an automatically generated e-mai

Review Request 59077: SENTRY-1739 Sentry Kafka tests do not stop periodic update after the test end

/apache/sentry/tests/e2e/kafka/AbstractKafkaSentryTestBase.java 5807f68128258cbf187eb13c518218d0e3f553aa Diff: https://reviews.apache.org/r/59077/diff/1/ Testing --- Thanks, Alexander Kolbasov

Re: SQL changes for SENTRY-1726

tes from another threads because we are using repeatable-read transaction isolation level. - Alex > > -Kalyan > > On Mon, May 8, 2017 at 7:03 PM, Alexander Kolbasov > wrote: > >> The major problem with read-modify-write approach is that it is difficult >> to

Re: Review Request 58727: SENTRY-1708 Extend the current test classes to handle multiple sentry servers

s://reviews.apache.org/r/58727/#comment247393> Note that even if you reuse existing configs, the new server may be started on a different port. - Alexander Kolbasov On May 5, 2017, 6:52 p.m., kalyan kumar kalvagadda wrote: > >

Re: Review Request 59031: SENTRY-1714 MetastorePlugin.java should quetly return from renameAuthzObject() when both paths are null

/sentry/binding/metastore/SentryMetastorePostEventListener.java Lines 416 (patched) <https://reviews.apache.org/r/59031/#comment247397> What should happen of one of them is null - either old path or new path? - Alexander Kolbasov On May 5, 2017, 9:38 p.m., kalyan kumar kalvagadda

Re: Review Request 58975: SENTRY-1669 HMSFollower should read current processed notification ID from database every time it runs

it doesn't compile properly. - Alexander Kolbasov On May 10, 2017, 12:34 a.m., kalyan kumar kalvagadda wrote: > > --- > This is an automatically generated e-mail. To reply, visit: > https://reviews.a

Review Request 59132: SENTRY-1759 UpdatableCache leaks connections

/UpdatableCache.java f272630171763b8afa2cbd826cb2004532e52a4a Diff: https://reviews.apache.org/r/59132/diff/1/ Testing --- Thanks, Alexander Kolbasov

Re: Review Request 59132: SENTRY-1759 UpdatableCache leaks connections

is involves more changes. - Alexander --- This is an automatically generated e-mail. To reply, visit: https://reviews.apache.org/r/59132/#review174482 --- On May 10, 20

Re: Review Request 59132: SENTRY-1759 UpdatableCache leaks connections

NTRY-1580. - Alexander --- This is an automatically generated e-mail. To reply, visit: https://reviews.apache.org/r/59132/#review174495 --- On May 10, 2017, 7:48 a.m.,

Re: Review Request 58808: SENTRY-1726 sql changes to store last notification-id processed

other upgrade scripts or not? - Alexander Kolbasov On May 9, 2017, 11:15 p.m., kalyan kumar kalvagadda wrote: > > --- > This is an automatically generated e-mail. To reply, visit: > https://reviews.apache

Re: Review Request 59120: SENTRY-1757: Avoid using local hive meta store using wrong configuration

2) In tests this is totally Ok but we do not want to print this every time run() is executed. Is there some way to distinguish between these two cases? - Alexander Kolbasov On May 10, 2017, 6:44 p.m., Na Li wrote: > > ---

Re: Review Request 58975: SENTRY-1669 HMSFollower should read current processed notification ID from database every time it runs

rom SentrySTore but here suddenly we assign it from the eventId which looks wrong. - Alexander Kolbasov On May 10, 2017, 9:45 p.m., kalyan kumar kalvagadda wrote: > > --- > This is an automa

Re: Review Request 59077: SENTRY-1739 Sentry Kafka tests do not stop periodic update after the test end

https://reviews.apache.org/r/59077/#review174491 --- On May 9, 2017, 12:13 a.m., Alexander Kolbasov wrote: > > --- > This is an automatically generated e-m

Re: Review Request 59077: SENTRY-1739 Sentry Kafka tests do not stop periodic update after the test end

It has to be because we must access it from static method. We don't have access to the specific instance. We rely on the fact that there is a single instance. - Alexander --- This is an automatically generated e-mail. To reply,

Re: Review Request 59077: SENTRY-1739 Sentry Kafka tests do not stop periodic update after the test end

This is an automatically generated e-mail. To reply, visit: https://reviews.apache.org/r/59077/#review174488 --- On May 9, 2017, 12:13 a.m., Alexander Kolbasov wrote: > > --- >

Re: Review Request 59077: SENTRY-1739 Sentry Kafka tests do not stop periodic update after the test end

> > When is "initialized" set to false? > > Alexander Kolbasov wrote: > That's the default initialization, but can make it explicit. It is set to false when it is initialized private boolean initialized = false; - Alexander -

Review Request 59167: SENTRY-1580: Provide pooled client connection model with HA

/TestAuthorize.java 0b1ef68a55a1d40637f771759735fc0838525a40 sentry-tests/sentry-tests-sqoop/src/test/java/org/apache/sentry/tests/e2e/sqoop/AbstractSqoopSentryTestBase.java 8a01e1c5c4ec58eded5621ad94ee28bc68d66c0b Diff: https://reviews.apache.org/r/59167/diff/1/ Testing --- Thanks, Alexander

Re: Review Request 59167: SENTRY-1580: Provide pooled client connection model with HA

8a01e1c5c4ec58eded5621ad94ee28bc68d66c0b Diff: https://reviews.apache.org/r/59167/diff/2/ Changes: https://reviews.apache.org/r/59167/diff/1-2/ Testing --- Thanks, Alexander Kolbasov

Re: [DISCUSS] Having an interactive chat account for Apache Sentry

As far as I know Kudu isn’t using Apache infrastructure, they are using private account and Slack integration: https://github.com/rauchg/slackin I think I can have a free private account on Slack with some reasonable limitations if that’s ok for others. I am

Re: Review Request 59204: SENTRY-1717: Sentry should emit log messages when it is ready to serve requests

--- This is an automatically generated e-mail. To reply, visit: https://reviews.apache.org/r/59204/#review174759 --- Ship it! Ship It! - Alexander Kolbasov On May 11, 2017, 8

Re: Review Request 59167: SENTRY-1580: Provide pooled client connection model with HA

sentry/core/common/transport/SentryTransportPool.java > > Lines 250 (patched) > > <https://reviews.apache.org/r/59167/diff/2/?file=1715191#file1715191line250> > > > > nit. typo fixed - Alexander --- This is an automatically generated e-mail. To reply, visit: htt

Re: Review Request 59167: SENTRY-1580: Provide pooled client connection model with HA

8a01e1c5c4ec58eded5621ad94ee28bc68d66c0b Diff: https://reviews.apache.org/r/59167/diff/3/ Changes: https://reviews.apache.org/r/59167/diff/2-3/ Testing --- Thanks, Alexander Kolbasov

Re: Review Request 59212: SENTRY-1763 Fix the config string for server load balancing

--- This is an automatically generated e-mail. To reply, visit: https://reviews.apache.org/r/59212/#review174814 --- Ship it! Ship It! - Alexander Kolbasov On May 11, 2017, 11

Review Request 59235: SENTRY-1765 CounterWait.update throw exception when input is old

/sentry/service/thrift/CounterWait.java 2b4ee8411a9d0e170cac44bd0cf9a560297412f2 Diff: https://reviews.apache.org/r/59235/diff/1/ Testing --- Thanks, Alexander Kolbasov

Re: Review Request 58975: SENTRY-1669 HMSFollower should read current processed notification ID from database every time it runs

/service/persistent/TestSentryStore.java Lines 2471 (patched) <https://reviews.apache.org/r/58975/#comment248095> I don't see how it can be null - Alexander Kolbasov On May 12, 2017, 5:38 p.m., kalyan kumar kalvagadda wrote: > >

Review Request 59252: SENTRY-1758 Improve Sentry memory usage by interning object names

ff8830f11a4f53801babbc0bb5deb386cfcbab42 Diff: https://reviews.apache.org/r/59252/diff/1/ Testing --- Thanks, Alexander Kolbasov

Re: Review Request 59241: SENTRY-1705: Do not start HMSFollower if Hive isn't configured

nown object state (e.g. hmsFollower being null). - Alexander Kolbasov On May 12, 2017, 7:49 p.m., Na Li wrote: > > --- > This is an automatically generated e-mail. To reply, visit: >

Re: Review Request 59252: SENTRY-1758 Improve Sentry memory usage by interning object names

/main/java/org/apache/sentry/provider/db/service/model/MSentryVersion.java ff8830f11a4f53801babbc0bb5deb386cfcbab42 Diff: https://reviews.apache.org/r/59252/diff/2/ Changes: https://reviews.apache.org/r/59252/diff/1-2/ Testing --- Thanks, Alexander Kolbasov

Re: Review Request 59252: SENTRY-1758 Improve Sentry memory usage by interning object names

his is an automatically generated e-mail. To reply, visit: https://reviews.apache.org/r/59252/#review174878 --- On May 13, 2017, 1:48 a.m., Alexander Kolbasov wrote: > > --- >

Re: Review Request 59167: SENTRY-1580: Provide pooled client connection model with HA

(updated) --- I used the interactive shell to test failover across two different servers. Thanks, Alexander Kolbasov

Re: Review Request 59167: SENTRY-1580: Provide pooled client connection model with HA

failover across two different servers. Thanks, Alexander Kolbasov

Re: Review Request 59274: SENTRY-1760: HMSFollower should detect when a full snapshot from HMS is required

vents().size() == expectedNotificationsSize) is false. - Alexander Kolbasov On May 15, 2017, 2:58 p.m., Sergio Pena wrote: > > --- > This is an automatically generated e-mail. To reply, visit: > https://re

Re: Review Request 59274: SENTRY-1760: HMSFollower should detect when a full snapshot from HMS is required

> On May 15, 2017, 3:14 p.m., Alexander Kolbasov wrote: > > sentry-provider/sentry-provider-db/src/main/java/org/apache/sentry/provider/db/service/persistent/SentryStore.java > > Lines 2513 (patched) > > <https://reviews.apache.org/r/59274/diff/1/?file=171

Re: Review Request 59274: SENTRY-1760: HMSFollower should detect when a full snapshot from HMS is required

that would cause this to happen? Sergio - what is the speci for notifications - does it guarantee that it returns *all* events since the specified value or *some number of consequitive events*"? - Alexander Kolbasov On May 15, 2017, 2:58 p.

Re: Review Request 58975: SENTRY-1669 HMSFollower should read current processed notification ID from database every time it runs

er then just save the ID and return? sentry-provider/sentry-provider-db/src/main/java/org/apache/sentry/service/thrift/HMSFollower.java Lines 600 (patched) <https://reviews.apache.org/r/58975/#comment248374> Why is this the case? If we are the leader we should process the rest, but

Re: Review Request 59120: SENTRY-1757: Avoid using local hive meta store using wrong configuration

.java Lines 786 (patched) <https://reviews.apache.org/r/59120/#comment248394> I don't understand this comment. Is there some async cleanup that is still going on? Why do you need to wait here? The cleanup should be done once we exit from this function. Or this isn't the case

Re: Review Request 59280: SENTRY-1751 HMSFollower should not persist empty full snapshot

--- This is an automatically generated e-mail. To reply, visit: https://reviews.apache.org/r/59280/#review175058 --- Ship it! Ship It! - Alexander Kolbasov On May 15, 2017, 6

Re: Review Request 59282: SENTRY-1716: HMSFollower doesn't need to save path info when HDFS sync is disabled

--- This is an automatically generated e-mail. To reply, visit: https://reviews.apache.org/r/59282/#review175068 --- Ship it! Ship It! - Alexander Kolbasov On May 15, 2017, 8

Re: Review Request 59274: SENTRY-1760: HMSFollower should detect when a full snapshot from HMS is required

weird scenario. - Alexander Kolbasov On May 15, 2017, 2:58 p.m., Sergio Pena wrote: > > --- > This is an automatically generated e-mail. To reply, visit: > https://reviews.

Re: Review Request 58975: SENTRY-1669 HMSFollower should read current processed notification ID from database every time it runs

istent/DeltaTransactionBlock.java Lines 91 (patched) <https://reviews.apache.org/r/58975/#comment248577> Nit - casting to PathUpdate isn't needed - Alexander Kolbasov On May 16, 2017, 5:11 p.m., kalyan kumar kalvagadda wrote: > > --

Re: Review Request 59167: SENTRY-1580: Provide pooled client connection model with HA

sentry/core/common/transport/TTransportWrapper.java > > Lines 79 (patched) > > <https://reviews.apache.org/r/59167/diff/4/?file=1717927#file1717927line79> > > > > Should we also put the class name in the string? Do you think it is useful? IMO it will o

Re: Review Request 59167: SENTRY-1580: Provide pooled client connection model with HA

t; sentry-provider/sentry-provider-db/src/main/java/org/apache/sentry/service/thrift/SentryServiceClientFactory.java > > Lines 66 (patched) > > <https://reviews.apache.org/r/59167/diff/4/?file=1717948#file1717948line67> > > > > As we do not have specific TransportF

Re: Review Request 59167: SENTRY-1580: Provide pooled client connection model with HA

--- This is an automatically generated e-mail. To reply, visit: https://reviews.apache.org/r/59167/#review175162 --- On May 13, 2017, 10:01 p.m., Alexander Kolbasov wrote: > > ---

Re: Review Request 59167: SENTRY-1580: Provide pooled client connection model with HA

shell to test failover across two different servers. Thanks, Alexander Kolbasov

Re: Review Request 59167: SENTRY-1580: Provide pooled client connection model with HA

- This is an automatically generated e-mail. To reply, visit: https://reviews.apache.org/r/59167/#review175162 ------- On May 17, 2017, 4:42 a.m., Alexander Kolbasov wrote: > > --

Review Request 59330: SENTRY-1768 Avoid detaching object on transaction exit when it isn't needed

https://reviews.apache.org/r/59330/diff/1/ Testing --- Thanks, Alexander Kolbasov

Re: Review Request 59330: SENTRY-1768 Avoid detaching object on transaction exit when it isn't needed

er ------- This is an automatically generated e-mail. To reply, visit: https://reviews.apache.org/r/59330/#review175250 --- On May 17, 2017, 7:37 a.m., Alexander Kolbasov wrote: &g

Review Request 59350: SENTRY-1743: Two SentrySTore instances are one too many

e9d9eec3526f2894199e81a253709dc3ed380b5b sentry-tests/sentry-tests-sqoop/src/test/java/org/apache/sentry/tests/e2e/sqoop/AbstractSqoopSentryTestBase.java 67de5acb36797b23261e5023c308222ad3344ea7 Diff: https://reviews.apache.org/r/59350/diff/1/ Testing --- Thanks, Alexander Kolbasov

Re: Review Request 59120: SENTRY-1757: Avoid using local hive meta store using wrong configuration

--- This is an automatically generated e-mail. To reply, visit: https://reviews.apache.org/r/59120/#review175304 --- Ship it! Ship It! - Alexander Kolbasov On May 17, 2017, 2

Re: Review Request 59350: SENTRY-1743: Two SentrySTore instances are one too many

2 --- On May 17, 2017, 7:19 p.m., Alexander Kolbasov wrote: > > --- > This is an automatically generated e-mail. To reply, visit: > https://reviews.apache.org/r/59350/ > -

Re: [DISCUSS] Having an interactive chat account for Apache Sentry

> > Btw, I'm not aware of any Apache infra for chat accounts. There is no “Apache hi-chat” account? - Alex > > On Thu, May 11, 2017 at 6:26 PM, Alexander Kolbasov > wrote: > >> As far as I know Kudu isn’t using Apache infrastructure, they are using >> priva

Re: [DISCUSS] Having an interactive chat account for Apache Sentry

017, at 4:26 PM, Alexander Kolbasov wrote: > > As far as I know Kudu isn’t using Apache infrastructure, they are using > private account and Slack integration: https://github.com/rauchg/slackin > <https://github.com/rauchg/slackin> > > I think I can have a free private

Review Request 59356: SENTRY-1770 Avoid more detaches on commit

/1/ Testing --- Thanks, Alexander Kolbasov

Re: Review Request 59317: SENTRY-1736 Generic service client should support Kerberos

py to use for modifications. - Alexander Kolbasov On May 18, 2017, 12:06 a.m., kalyan kumar kalvagadda wrote: > > --- > This is an automatically generated e-mail. To reply, visit:

Review Request 59361: SENTRY-1772 Permissions created before table creation are not reflected in HDFS ACLs

/main/java/org/apache/sentry/service/thrift/HMSFollower.java 5dc0ad338bbfee37a537bdebbd3a24e76714b796 Diff: https://reviews.apache.org/r/59361/diff/1/ Testing --- Thanks, Alexander Kolbasov

Re: [DISCUSS] Having an interactive chat account for Apache Sentry

May 2017, at 12:59 am, Alexander Kolbasov > wrote: > > > > So far I only saw comments from people working at Cloudera. We do have > tools to talk to each other within the company - so far it isn’t clear > whether this is beneficial for others as well. It would be good to hear &

Re: Review Request 59362: SENTRY-1235 - Some pom changes

--- This is an automatically generated e-mail. To reply, visit: https://reviews.apache.org/r/59362/#review175379 --- Ship it! Ship It! - Alexander Kolbasov On May 18, 2017, 9

Re: Review Request 59317: SENTRY-1736 Generic service client should support Kerberos

--- This is an automatically generated e-mail. To reply, visit: https://reviews.apache.org/r/59317/#review175448 --- Ship it! Ship It! - Alexander Kolbasov On May 18, 2017, 10

Re: Review Request 59241: SENTRY-1705: Do not start HMSFollower if Hive isn't configured

-provider-db/src/main/java/org/apache/sentry/service/thrift/SentryService.java Lines 324 (patched) <https://reviews.apache.org/r/59241/#comment248903> So we no longer have the situation where metastore URI isn't present initially but later becomes available? - Alexander Kolbasov

Re: [DISCUSS] Having an interactive chat account for Apache Sentry

gt; > -Rams > >> On 18 May 2017, at 8:57 am, Alexander Kolbasov wrote: >> >> Rams, do you have any preference - IRC vs Slack? >> >> On Thu, May 18, 2017 at 12:11 AM Rams Venkatesh >> wrote: >> >>> I would be very interested having chat acc

Re: [DISCUSS] Having an interactive chat account for Apache Sentry

Sent! On Thu, May 18, 2017 at 5:25 PM Nishith Nand wrote: > Hi Alex, > > I would like and invite to the slack channel too. > > Thanks, > Nishith > > > > On May 18, 2017, at 5:20 PM, Alexander Kolbasov > wrote: > > > > I’ve set up an experimental Sla

Re: Review Request 59361: SENTRY-1772 Permissions created before table creation are not reflected in HDFS ACLs

/1-2/ Testing --- Thanks, Alexander Kolbasov

Re: Review Request 58975: SENTRY-1669 HMSFollower should read current processed notification ID from database every time it runs

--- This is an automatically generated e-mail. To reply, visit: https://reviews.apache.org/r/58975/#review175471 --- Ship it! Ship It! - Alexander Kolbasov On May 18, 2017, 3

Re: Review Request 59241: SENTRY-1705: Do not start HMSFollower if Hive isn't configured

--- This is an automatically generated e-mail. To reply, visit: https://reviews.apache.org/r/59241/#review175472 --- Ship it! Ship It! - Alexander Kolbasov On May 19, 2017, 4

Re: Review Request 59397: SENTRY-1776 SentryStore should clear SENTRY_HMS_NOTIFICATION_ID while clearing store

--- This is an automatically generated e-mail. To reply, visit: https://reviews.apache.org/r/59397/#review175506 --- Ship it! Ship It! - Alexander Kolbasov On May 19, 2017, 1

Re: Review Request 59414: SENTRY-1771: HDFS client requests full permission update multiple times.

---- > > (Updated May 19, 2017, 6:56 p.m.) > > > Review request for sentry, Alexander Kolbasov, Hao Hao, kalyan kumar > kalvagadda, Sergio Pena, and Vamsee Yarlagadda. > > > Bugs: SENTRY-1771 > https://issues.apache.org/ji

Re: Review Request 59424: SENTRY-1736 Generic service client should support Kerberos(Continuation Fix)

another JIRA for subsequent changes. - Alexander Kolbasov On May 19, 2017, 11:27 p.m., kalyan kumar kalvagadda wrote: > > --- > This is an automatically generated e-mail. To reply, visit: > https://reviews.apache

Re: Review Request 59167: SENTRY-1580: Provide pooled client connection model with HA

h will cause a retry on the next attempt. - Alexander --- This is an automatically generated e-mail. To reply, visit: https://reviews.apache.org/r/59167/#review175427 -------

Re: Review Request 59167: SENTRY-1580: Provide pooled client connection model with HA

eviews.apache.org/r/59167/#review175433 --- On May 17, 2017, 4:42 a.m., Alexander Kolbasov wrote: > > --- > This is an automatically generated e-mail. To re

Information about 1.7 release is missing from the wiki

I noticed that the wiki at https://sentry.apache.org/general/history.html is missing information about 1.7 release. Would be great to add it there. - Alex

Who maintains sentry.apache.org?

Who has write access to https://sentry.apache.org/index.html? How is it usually updated? - Alex

Re: Review Request 58896: SENTRY-1744 Simplify creation of DelegateSentryStore

n on the JIRA - I'll put my comment there. - Alexander --- This is an automatically generated e-mail. To reply, visit: https://reviews.apache.org/r/58896/#review174492 ----

Re: Review Request 59350: SENTRY-1743: Two SentrySTore instances are one too many

e not set in which case instance.get() will return null. This code is updated a bit in the new patch. - Alexander --- This is an automatically generated e-mail. To reply, visit: https://reviews.apache.org/r/59350/#review17530

Re: Review Request 59167: SENTRY-1580: Provide pooled client connection model with HA

8a01e1c5c4ec58eded5621ad94ee28bc68d66c0b Diff: https://reviews.apache.org/r/59167/diff/6/ Changes: https://reviews.apache.org/r/59167/diff/5-6/ Testing --- I used the interactive shell to test failover across two different servers. Thanks, Alexander Kolbasov

[DISCUSS] Considering Sentry release

I would like to start a preliminary discussion about the new Sentry release. To make this interesting, I suggest to consider thinking about two releases. 1) There are a lot of accumulated fixes since 1.7. IMO it would be useful to release those as 1.8 release. 2) Sentry HA includes a large numb

Re: Review Request 59167: SENTRY-1580: Provide pooled client connection model with HA

iew175920 --- On May 24, 2017, 7:49 a.m., Alexander Kolbasov wrote: > > --- > This is an automatically generated e-mail. To reply, visit: > https://revie

Re: Review Request 59510: SENTRY-1774 HMSFollower should always depend on persisted information to decide is full snapshot is needed

ovider-db/src/main/java/org/apache/sentry/service/thrift/HMSFollower.java Line 103 (original), 101 (patched) <https://reviews.apache.org/r/59510/#comment249330> This return is redundand now - Alexander Kolbasov On May 24, 2017, 1:36 a.m., kalya

Review Request 59567: SENTRY-1783 alterSentryRoleGrantPrivilegeCore does more persistence work than required

-server/src/main/java/org/apache/sentry/provider/db/service/persistent/SentryStore.java f0d2a6b3a74d8d88976aeec32bba901d02e8329f Diff: https://reviews.apache.org/r/59567/diff/1/ Testing --- Thanks, Alexander Kolbasov

Re: Review Request 59576: SENTRY-1785 Fix TestKrbConnectionTimeout test

-provider-db/src/test/java/org/apache/sentry/service/thrift/SentryServiceIntegrationBase.java Lines 204 (patched) <https://reviews.apache.org/r/59576/#comment249430> Why do we need to set this property to true? Can we just remove the assert that checks for it? - Alexander Kolbasov On May 25

Re: Review Request 59576: SENTRY-1785 Fix TestKrbConnectionTimeout test

--- This is an automatically generated e-mail. To reply, visit: https://reviews.apache.org/r/59576/#review176150 --- Ship it! Ship It! - Alexander Kolbasov On May 25, 2017, 10

Re: Review Request 59167: SENTRY-1580: Provide pooled client connection model with HA

, Alexander Kolbasov

Re: [DISCUSS] Considering Sentry release

t; > The use of Hive notifications logs requires us to support Hive 2.x > > > > (probably next 2.3 release), and it would be interesting to start the > > > > support of Hive Authorization V2 API that brings more > fine-granularity > > > > p

Re: Review Request 59510: SENTRY-1774 HMSFollower should always depend on persisted information to decide is full snapshot is needed

--- This is an automatically generated e-mail. To reply, visit: https://reviews.apache.org/r/59510/#review176227 --- Ship it! Ship It! - Alexander Kolbasov On May 25, 2017, 1

Re: Review Request 59510: SENTRY-1774 HMSFollower should always depend on persisted information to decide is full snapshot is needed

! sentry-provider/sentry-provider-db/src/main/java/org/apache/sentry/provider/db/service/persistent/SentryStore.java Lines 2891 (patched) <https://reviews.apache.org/r/59510/#comment249575> Please camelize the name to isTableEmpty - Alexander Kolbasov On May 26, 2017, 9:30 p.m., kalyan

Re: Review Request 59655: SENTRY-1788 Sentry Store may use JDO object after the associated data is removed in DB

pache/sentry/provider/db/service/persistent/SentryStore.java Line 1745 (original), 1751 (patched) <https://reviews.apache.org/r/59655/#comment249884> Here and below, newTPrivilege never changes, so we can pre-compute everything upfront and simpli

Re: Review Request 59655: SENTRY-1788 Sentry Store may use JDO object after the associated data is removed in DB

this function better. - Alexander Kolbasov On May 30, 2017, 11:17 p.m., kalyan kumar kalvagadda wrote: > > --- > This is an automatically generated e-mail. To reply, visit: > https:/

Re: Review Request 59350: SENTRY-1743: Two SentrySTore instances are one too many

e master branch or sentry-ha-redesign branch? - Alexander --- This is an automatically generated e-mail. To reply, visit: https://reviews.apache.org/r/59350/#review175311 ------

Re: Review Request 59350: SENTRY-1743: Two SentrySTore instances are one too many

singleton class? How about renaming it to getSentryStore() - Alexander --- This is an automatically generated e-mail. To reply, visit: https://reviews.apache.org/r/59350/#review176501 -----

Re: Review Request 59414: SENTRY-1771: HDFS client requests full permission update multiple times.

this fix? - Alexander Kolbasov On May 19, 2017, 6:56 p.m., Na Li wrote: > > --- > This is an automatically generated e-mail. To reply, visit: > https://reviews.apache

Re: Review Request 59350: SENTRY-1743: Two SentrySTore instances are one too many

line134> > > > > The method name does not match with what it is return. Some people may > > see newSentryStore() as a method that creates a new SentryStore every time > > is called, but this creates it only once. Should we rename the method? > > Perhaps have a

Re: Review Request 59655: SENTRY-1788 Sentry Store may use JDO object after the associated data is removed in DB

JIRA so that we don't forget them. - Alexander Kolbasov On May 31, 2017, 8:30 p.m., kalyan kumar kalvagadda wrote: > > --- > This is an automatically generated e-mail. To reply, visit: > https://reviews.a

Re: Review Request 59655: SENTRY-1788 Sentry Store may use JDO object after the associated data is removed in DB

--- This is an automatically generated e-mail. To reply, visit: https://reviews.apache.org/r/59655/#review176550 --- - Alexander Kolbasov On May 31, 2017, 8:30 p.m., kalyan kumar

Updating wiki for old Sentry HA

We have an old page on wiki for configuring Sentry ha - https://cwiki.apache.org/confluence/display/SENTRY/Sentry+HA+Server+Client+Configuration . I'd like to either remove it or mark it as obsolete. Any objections?

Re: Review Request 59566: SENTRY-1784: DBUpdateForwarder returns empty update list to HDFS instead of full update

ts in the code? - Alexander Kolbasov On June 1, 2017, 3:17 a.m., Na Li wrote: > > --- > This is an automatically generated e-mail. To reply, visit: > https://reviews.a

Re: Updating wiki for old Sentry HA

t; option? > > > > On Thu, Jun 1, 2017 at 5:13 AM, Rams Venkatesh > > wrote: > > > > > Obsolete is more appropriate than completely removing it. > > > > > > -Rams > > > > > > > On 1 Jun 2017, at 10:07 am, Alexander Kolbasov

Re: Updating wiki for old Sentry HA

Sounds reasonable to me. On Thu, Jun 1, 2017 at 10:47 AM, Rams Venkatesh wrote: > We can mark as obsolete till the current HA is released and then remove it > completely. > > -Rams > > > On 1 Jun 2017, at 6:34 pm, Alexander Kolbasov > wrote: > > > > What

Re: Updating wiki for old Sentry HA

Hmm, seems like I can't edit the page - do I need special permissions for that? On Thu, Jun 1, 2017 at 12:16 PM, Alexander Kolbasov wrote: > Sounds reasonable to me. > > On Thu, Jun 1, 2017 at 10:47 AM, Rams Venkatesh > wrote: > >> We can mark as obsolete till the

Re: Updating wiki for old Sentry HA

;t edit it either. Who is the admin of the Sentry > > confluence site? If that person is no longer active, then I guess > Alexander > > you could file a JIRA with INFRA to be made an admin as PMC chair. > > > > Colm. > > > > On Fri, Jun 2, 2017 at 2:44 AM, Alexa

<    3   4   5   6   7   8   9   10   >