Mailing-List: contact issues-help@hbase.apache.org; run by ezmlm
Precedence: bulk
Date: Wed, 1 Mar 2017 23:18:45 +0000 (UTC)
From: "Andrew Purtell (JIRA)" <jira@apache.org>
To: issues@hbase.apache.org
Message-ID: <JIRA.13047426.1488397139000.36966.1488410325470@Atlassian.JIRA>
In-Reply-To: <JIRA.13047426.1488397139000@Atlassian.JIRA>
References: <JIRA.13047426.1488397139000@Atlassian.JIRA> <JIRA.13047426.1488397139725@jira-lw-us.apache.org>
Subject: [jira] [Comment Edited] (HBASE-17714) Client heartbeats seems to be
 broken
MIME-Version: 1.0
Content-Type: text/plain; charset=utf-8
Content-Transfer-Encoding: 7bit
archived-at: Wed, 01 Mar 2017 23:18:53 -0000


    [ https://issues.apache.org/jira/browse/HBASE-17714?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15891268#comment-15891268 ] 

Andrew Purtell edited comment on HBASE-17714 at 3/1/17 11:18 PM:
-----------------------------------------------------------------

e5c1a80 (HBASE-15645 hbase.rpc.timeout is not used in operations of HTable) is the commit that causes RenewLeaseIT to start failing. You have to apply HBASE-16420 (Fix source incompatibility of Table interface) after checking out e5c1a80 to get something that 4.x-HBase-1.1 will compile against.  

HBASE-15645 was shipped in 1.1.5, 1.2.2, and 1.3.0.  

What this change does is fix where the client was not actually honoring RPC timeouts prior to the change. [~samarthjain] are you sure RenewLeaseIT actually renews the lease or allows for a client heartbeat to happen before the RPC times out? The test sets a very short RPC timeout (2000ms) but makes no other configuration changes


was (Author: apurtell):
e5c1a80 (HBASE-15645 hbase.rpc.timeout is not used in operations of HTable) is the commit that causes RenewLeaseIT to start failing. You have to apply HBASE-16420 (Fix source incompatibility of Table interface) after checking out e5c1a80 to get something that 4.x-HBase-1.1 will compile against.  

HBASE-15645 was shipped in 1.1.5, 1.2.2, and 1.3.0.  

What this change does is fix where the client was not actually honoring RPC timeouts prior to the change. [~samarthjain] are you sure RenewLeaseIT actually renews the lease before the RPC times out? The test sets a very short RPC timeout (2000ms)


> Client heartbeats seems to be broken
> ------------------------------------
>
>                 Key: HBASE-17714
>                 URL: https://issues.apache.org/jira/browse/HBASE-17714
>             Project: HBase
>          Issue Type: Bug
>            Reporter: Samarth Jain
>
> We have a test in Phoenix where we introduce an artificial sleep of 2 times the RPC timeout in preScannerNext() hook of a co-processor. 
> {code}
>  public static class SleepingRegionObserver extends SimpleRegionObserver {
>         public SleepingRegionObserver() {}
>         
>         @Override
>         public boolean preScannerNext(final ObserverContext<RegionCoprocessorEnvironment> c,
>                 final InternalScanner s, final List<Result> results,
>                 final int limit, final boolean hasMore) throws IOException {
>             try {
>                 if (SLEEP_NOW && c.getEnvironment().getRegion().getRegionInfo().getTable().getNameAsString().equals(TABLE_NAME)) {
>                     Thread.sleep(RPC_TIMEOUT * 2);
>                 }
>             } catch (InterruptedException e) {
>                 throw new IOException(e);
>             }
>             return super.preScannerNext(c, s, results, limit, hasMore);
>         }
>     }
> {code}
> This test was passing fine till 1.1.3 but started failing sometime before 1.1.9 with an OutOfOrderScannerException. See PHOENIX-3702. [~lhofhansl] mentioned that we have client heartbeats enabled and that should prevent us from running into issues like this. FYI, this test fails with 1.2.3 version of HBase too.
> CC [~apurtell], [~jamestaylor]


--
This message was sent by Atlassian JIRA
(v6.3.15#6346)