[jira] [Updated] (HDFS-7725) Incorrect "nodes in service" metrics caused all writes to fail
[
https://issues.apache.org/jira/browse/HDFS-7725?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Junping Du updated HDFS-7725:
-
Fix Version/s: 2.8.0
> Incorrect "nodes in service" metrics caused all writes to fail
> --
>
> Key: HDFS-7725
> URL: https://issues.apache.org/jira/browse/HDFS-7725
> Project: Hadoop HDFS
> Issue Type: Bug
>Reporter: Ming Ma
>Assignee: Ming Ma
> Labels: 2.7.2-candidate
> Fix For: 2.8.0, 2.7.2, 3.0.0-alpha1
>
> Attachments: HDFS-7725-2.patch, HDFS-7725-3.patch, HDFS-7725.patch
>
>
> One of our clusters sometimes couldn't allocate blocks from any DNs.
> BlockPlacementPolicyDefault complains with the following messages for all DNs.
> {noformat}
> the node is too busy (load:x > y)
> {noformat}
> It turns out the {{HeartbeatManager}}'s {{nodesInService}} was computed
> incorrectly when admins decomm or recomm dead nodes. Here are two scenarios.
> * Decomm dead nodes. It turns out HDFS-7374 has fixed it; not sure if it is
> intentional. cc / [~zhz], [~andrew.wang], [~atm] Here is the sequence of
> event without HDFS-7374.
> ** Cluster has one live node. nodesInService == 1
> ** The node becomes dead. nodesInService == 0
> ** Decomm the node. nodesInService == -1
> * However, HDFS-7374 introduces another inconsistency when recomm is involved.
> ** Cluster has one live node. nodesInService == 1
> ** The node becomes dead. nodesInService == 0
> ** Decomm the node. nodesInService == 0
> ** Recomm the node. nodesInService == 1
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
-
To unsubscribe, e-mail: hdfs-issues-unsubscr...@hadoop.apache.org
For additional commands, e-mail: hdfs-issues-h...@hadoop.apache.org
[jira] [Updated] (HDFS-7725) Incorrect "nodes in service" metrics caused all writes to fail
[
https://issues.apache.org/jira/browse/HDFS-7725?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Arpit Agarwal updated HDFS-7725:
Fix Version/s: (was: 3.0.0)
> Incorrect "nodes in service" metrics caused all writes to fail
> --
>
> Key: HDFS-7725
> URL: https://issues.apache.org/jira/browse/HDFS-7725
> Project: Hadoop HDFS
> Issue Type: Bug
>Reporter: Ming Ma
>Assignee: Ming Ma
> Labels: 2.7.2-candidate
> Fix For: 2.7.2
>
> Attachments: HDFS-7725-2.patch, HDFS-7725-3.patch, HDFS-7725.patch
>
>
> One of our clusters sometimes couldn't allocate blocks from any DNs.
> BlockPlacementPolicyDefault complains with the following messages for all DNs.
> {noformat}
> the node is too busy (load:x > y)
> {noformat}
> It turns out the {{HeartbeatManager}}'s {{nodesInService}} was computed
> incorrectly when admins decomm or recomm dead nodes. Here are two scenarios.
> * Decomm dead nodes. It turns out HDFS-7374 has fixed it; not sure if it is
> intentional. cc / [~zhz], [~andrew.wang], [~atm] Here is the sequence of
> event without HDFS-7374.
> ** Cluster has one live node. nodesInService == 1
> ** The node becomes dead. nodesInService == 0
> ** Decomm the node. nodesInService == -1
> * However, HDFS-7374 introduces another inconsistency when recomm is involved.
> ** Cluster has one live node. nodesInService == 1
> ** The node becomes dead. nodesInService == 0
> ** Decomm the node. nodesInService == 0
> ** Recomm the node. nodesInService == 1
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
[jira] [Updated] (HDFS-7725) Incorrect "nodes in service" metrics caused all writes to fail
[
https://issues.apache.org/jira/browse/HDFS-7725?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Kihwal Lee updated HDFS-7725:
-
Fix Version/s: (was: 2.8.0)
2.7.2
3.0.0
> Incorrect "nodes in service" metrics caused all writes to fail
> --
>
> Key: HDFS-7725
> URL: https://issues.apache.org/jira/browse/HDFS-7725
> Project: Hadoop HDFS
> Issue Type: Bug
>Reporter: Ming Ma
>Assignee: Ming Ma
> Labels: 2.7.2-candidate
> Fix For: 3.0.0, 2.7.2
>
> Attachments: HDFS-7725-2.patch, HDFS-7725-3.patch, HDFS-7725.patch
>
>
> One of our clusters sometimes couldn't allocate blocks from any DNs.
> BlockPlacementPolicyDefault complains with the following messages for all DNs.
> {noformat}
> the node is too busy (load:x > y)
> {noformat}
> It turns out the {{HeartbeatManager}}'s {{nodesInService}} was computed
> incorrectly when admins decomm or recomm dead nodes. Here are two scenarios.
> * Decomm dead nodes. It turns out HDFS-7374 has fixed it; not sure if it is
> intentional. cc / [~zhz], [~andrew.wang], [~atm] Here is the sequence of
> event without HDFS-7374.
> ** Cluster has one live node. nodesInService == 1
> ** The node becomes dead. nodesInService == 0
> ** Decomm the node. nodesInService == -1
> * However, HDFS-7374 introduces another inconsistency when recomm is involved.
> ** Cluster has one live node. nodesInService == 1
> ** The node becomes dead. nodesInService == 0
> ** Decomm the node. nodesInService == 0
> ** Recomm the node. nodesInService == 1
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
[jira] [Updated] (HDFS-7725) Incorrect "nodes in service" metrics caused all writes to fail
[
https://issues.apache.org/jira/browse/HDFS-7725?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Kihwal Lee updated HDFS-7725:
-
Labels: 2.7.2-candidate (was: )
> Incorrect "nodes in service" metrics caused all writes to fail
> --
>
> Key: HDFS-7725
> URL: https://issues.apache.org/jira/browse/HDFS-7725
> Project: Hadoop HDFS
> Issue Type: Bug
>Reporter: Ming Ma
>Assignee: Ming Ma
> Labels: 2.7.2-candidate
> Fix For: 2.8.0
>
> Attachments: HDFS-7725-2.patch, HDFS-7725-3.patch, HDFS-7725.patch
>
>
> One of our clusters sometimes couldn't allocate blocks from any DNs.
> BlockPlacementPolicyDefault complains with the following messages for all DNs.
> {noformat}
> the node is too busy (load:x > y)
> {noformat}
> It turns out the {{HeartbeatManager}}'s {{nodesInService}} was computed
> incorrectly when admins decomm or recomm dead nodes. Here are two scenarios.
> * Decomm dead nodes. It turns out HDFS-7374 has fixed it; not sure if it is
> intentional. cc / [~zhz], [~andrew.wang], [~atm] Here is the sequence of
> event without HDFS-7374.
> ** Cluster has one live node. nodesInService == 1
> ** The node becomes dead. nodesInService == 0
> ** Decomm the node. nodesInService == -1
> * However, HDFS-7374 introduces another inconsistency when recomm is involved.
> ** Cluster has one live node. nodesInService == 1
> ** The node becomes dead. nodesInService == 0
> ** Decomm the node. nodesInService == 0
> ** Recomm the node. nodesInService == 1
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
[jira] [Updated] (HDFS-7725) Incorrect "nodes in service" metrics caused all writes to fail
[
https://issues.apache.org/jira/browse/HDFS-7725?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Andrew Wang updated HDFS-7725:
--
Resolution: Fixed
Fix Version/s: 2.8.0
Status: Resolved (was: Patch Available)
Committed to trunk and branch-2, thanks for the patch Ming, Zhe for also
reviewing.
The failed test also failed twice for me without the patch applied, it seems
independently broken.
> Incorrect "nodes in service" metrics caused all writes to fail
> --
>
> Key: HDFS-7725
> URL: https://issues.apache.org/jira/browse/HDFS-7725
> Project: Hadoop HDFS
> Issue Type: Bug
>Reporter: Ming Ma
>Assignee: Ming Ma
> Fix For: 2.8.0
>
> Attachments: HDFS-7725-2.patch, HDFS-7725-3.patch, HDFS-7725.patch
>
>
> One of our clusters sometimes couldn't allocate blocks from any DNs.
> BlockPlacementPolicyDefault complains with the following messages for all DNs.
> {noformat}
> the node is too busy (load:x > y)
> {noformat}
> It turns out the {{HeartbeatManager}}'s {{nodesInService}} was computed
> incorrectly when admins decomm or recomm dead nodes. Here are two scenarios.
> * Decomm dead nodes. It turns out HDFS-7374 has fixed it; not sure if it is
> intentional. cc / [~zhz], [~andrew.wang], [~atm] Here is the sequence of
> event without HDFS-7374.
> ** Cluster has one live node. nodesInService == 1
> ** The node becomes dead. nodesInService == 0
> ** Decomm the node. nodesInService == -1
> * However, HDFS-7374 introduces another inconsistency when recomm is involved.
> ** Cluster has one live node. nodesInService == 1
> ** The node becomes dead. nodesInService == 0
> ** Decomm the node. nodesInService == 0
> ** Recomm the node. nodesInService == 1
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
[jira] [Updated] (HDFS-7725) Incorrect "nodes in service" metrics caused all writes to fail
[
https://issues.apache.org/jira/browse/HDFS-7725?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Ming Ma updated HDFS-7725:
--
Attachment: HDFS-7725-3.patch
Thanks, Andrew and Zhe. The latest patch moves the liveness check to
{{HeartbeatManager}} with some minor changes to {{DecommissionManager}}'s
{{startDecommission}}. The patch also has the other suggestions you mentioned.
> Incorrect "nodes in service" metrics caused all writes to fail
> --
>
> Key: HDFS-7725
> URL: https://issues.apache.org/jira/browse/HDFS-7725
> Project: Hadoop HDFS
> Issue Type: Bug
>Reporter: Ming Ma
>Assignee: Ming Ma
> Attachments: HDFS-7725-2.patch, HDFS-7725-3.patch, HDFS-7725.patch
>
>
> One of our clusters sometimes couldn't allocate blocks from any DNs.
> BlockPlacementPolicyDefault complains with the following messages for all DNs.
> {noformat}
> the node is too busy (load:x > y)
> {noformat}
> It turns out the {{HeartbeatManager}}'s {{nodesInService}} was computed
> incorrectly when admins decomm or recomm dead nodes. Here are two scenarios.
> * Decomm dead nodes. It turns out HDFS-7374 has fixed it; not sure if it is
> intentional. cc / [~zhz], [~andrew.wang], [~atm] Here is the sequence of
> event without HDFS-7374.
> ** Cluster has one live node. nodesInService == 1
> ** The node becomes dead. nodesInService == 0
> ** Decomm the node. nodesInService == -1
> * However, HDFS-7374 introduces another inconsistency when recomm is involved.
> ** Cluster has one live node. nodesInService == 1
> ** The node becomes dead. nodesInService == 0
> ** Decomm the node. nodesInService == 0
> ** Recomm the node. nodesInService == 1
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
[jira] [Updated] (HDFS-7725) Incorrect "nodes in service" metrics caused all writes to fail
[
https://issues.apache.org/jira/browse/HDFS-7725?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Ming Ma updated HDFS-7725:
--
Attachment: HDFS-7725-2.patch
Thanks, Zhe. Here is the rebase of the patch.
> Incorrect "nodes in service" metrics caused all writes to fail
> --
>
> Key: HDFS-7725
> URL: https://issues.apache.org/jira/browse/HDFS-7725
> Project: Hadoop HDFS
> Issue Type: Bug
>Reporter: Ming Ma
>Assignee: Ming Ma
> Attachments: HDFS-7725-2.patch, HDFS-7725.patch
>
>
> One of our clusters sometimes couldn't allocate blocks from any DNs.
> BlockPlacementPolicyDefault complains with the following messages for all DNs.
> {noformat}
> the node is too busy (load:x > y)
> {noformat}
> It turns out the {{HeartbeatManager}}'s {{nodesInService}} was computed
> incorrectly when admins decomm or recomm dead nodes. Here are two scenarios.
> * Decomm dead nodes. It turns out HDFS-7374 has fixed it; not sure if it is
> intentional. cc / [~zhz], [~andrew.wang], [~atm] Here is the sequence of
> event without HDFS-7374.
> ** Cluster has one live node. nodesInService == 1
> ** The node becomes dead. nodesInService == 0
> ** Decomm the node. nodesInService == -1
> * However, HDFS-7374 introduces another inconsistency when recomm is involved.
> ** Cluster has one live node. nodesInService == 1
> ** The node becomes dead. nodesInService == 0
> ** Decomm the node. nodesInService == 0
> ** Recomm the node. nodesInService == 1
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
[jira] [Updated] (HDFS-7725) Incorrect "nodes in service" metrics caused all writes to fail
[
https://issues.apache.org/jira/browse/HDFS-7725?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Ming Ma updated HDFS-7725:
--
Description:
One of our clusters sometimes couldn't allocate blocks from any DNs.
BlockPlacementPolicyDefault complains with the following messages for all DNs.
{noformat}
the node is too busy (load:x > y)
{noformat}
It turns out the {{HeartbeatManager}}'s {{nodesInService}} was computed
incorrectly when admins decomm or recomm dead nodes. Here are two scenarios.
* Decomm dead nodes. It turns out HDFS-7374 has fixed it; not sure if it is
intentional. cc / [~zhz], [~andrew.wang], [~atm] Here is the sequence of event
without HDFS-7374.
** Cluster has one live node. nodesInService == 1
** The node becomes dead. nodesInService == 0
** Decomm the node. nodesInService == -1
* However, HDFS-7374 introduces another inconsistency when recomm is involved.
** Cluster has one live node. nodesInService == 1
** The node becomes dead. nodesInService == 0
** Decomm the node. nodesInService == 0
** Recomm the node. nodesInService == 1
was:
`One of our clusters sometimes couldn't allocate blocks from any DNs.
BlockPlacementPolicyDefault complains with the following messages for all DNs.
{noformat}
the node is too busy (load:x > y)
{noformat}
It turns out the {{HeartbeatManager}}'s {{nodesInService}} was computed
incorrectly when admins decomm or recomm dead nodes. Here are two scenarios.
* Decomm dead nodes. It turns out HDFS-7374 has fixed it; not sure if it is
intentional. cc / [~zhz], [~andrew.wang], [~atm] Here is the sequence of event
without HDFS-7374.
** Cluster has one live node. nodesInService == 1
** The node becomes dead. nodesInService == 0
** Decomm the node. nodesInService == -1
* However, HDFS-7374 introduces another inconsistency when recomm is involved.
** Cluster has one live node. nodesInService == 1
** The node becomes dead. nodesInService == 0
** Decomm the node. nodesInService == 0
** Recomm the node. nodesInService == 1
> Incorrect "nodes in service" metrics caused all writes to fail
> --
>
> Key: HDFS-7725
> URL: https://issues.apache.org/jira/browse/HDFS-7725
> Project: Hadoop HDFS
> Issue Type: Bug
>Reporter: Ming Ma
>Assignee: Ming Ma
> Attachments: HDFS-7725.patch
>
>
> One of our clusters sometimes couldn't allocate blocks from any DNs.
> BlockPlacementPolicyDefault complains with the following messages for all DNs.
> {noformat}
> the node is too busy (load:x > y)
> {noformat}
> It turns out the {{HeartbeatManager}}'s {{nodesInService}} was computed
> incorrectly when admins decomm or recomm dead nodes. Here are two scenarios.
> * Decomm dead nodes. It turns out HDFS-7374 has fixed it; not sure if it is
> intentional. cc / [~zhz], [~andrew.wang], [~atm] Here is the sequence of
> event without HDFS-7374.
> ** Cluster has one live node. nodesInService == 1
> ** The node becomes dead. nodesInService == 0
> ** Decomm the node. nodesInService == -1
> * However, HDFS-7374 introduces another inconsistency when recomm is involved.
> ** Cluster has one live node. nodesInService == 1
> ** The node becomes dead. nodesInService == 0
> ** Decomm the node. nodesInService == 0
> ** Recomm the node. nodesInService == 1
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
[jira] [Updated] (HDFS-7725) Incorrect "nodes in service" metrics caused all writes to fail
[
https://issues.apache.org/jira/browse/HDFS-7725?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Ming Ma updated HDFS-7725:
--
Description:
`One of our clusters sometimes couldn't allocate blocks from any DNs.
BlockPlacementPolicyDefault complains with the following messages for all DNs.
{noformat}
the node is too busy (load:x > y)
{noformat}
It turns out the {{HeartbeatManager}}'s {{nodesInService}} was computed
incorrectly when admins decomm or recomm dead nodes. Here are two scenarios.
* Decomm dead nodes. It turns out HDFS-7374 has fixed it; not sure if it is
intentional. cc / [~zhz], [~andrew.wang], [~atm] Here is the sequence of event
without HDFS-7374.
** Cluster has one live node. nodesInService == 1
** The node becomes dead. nodesInService == 0
** Decomm the node. nodesInService == -1
* However, HDFS-7374 introduces another inconsistency when recomm is involved.
** Cluster has one live node. nodesInService == 1
** The node becomes dead. nodesInService == 0
** Decomm the node. nodesInService == 0
** Recomm the node. nodesInService == 1
was:
One of our clusters sometimes couldn't allocate blocks from any DNs.
BlockPlacementPolicyDefault complains with the following messages for all DNs.
{noformat}
the node is too busy (load:x > y)
{noformat}
It turns out the {{HeartbeatManager}}'s {{nodesInService}} was computed
incorrectly when admins decomm or recomm dead nodes. Here are two scenarios.
* Decomm dead nodes. It turns out HDFS-7374 has fixed it; not sure if it is
intentional. cc / [~zhz], [~andrew.wang], [~atm] Here is the sequence of event
without HDFS-7374.
** Cluster has one live node. nodesInService == 1
** The node becomes dead. nodesInService == 0
** Decomm the node. nodesInService == -1
* However, HDFS-7374 introduces another inconsistency when recomm is involved.
** Cluster has one live node. nodesInService == 1
** The node becomes dead. nodesInService == 0
** Decomm the node. nodesInService == 0
** Recomm the node. nodesInService == 1
> Incorrect "nodes in service" metrics caused all writes to fail
> --
>
> Key: HDFS-7725
> URL: https://issues.apache.org/jira/browse/HDFS-7725
> Project: Hadoop HDFS
> Issue Type: Bug
>Reporter: Ming Ma
>Assignee: Ming Ma
> Attachments: HDFS-7725.patch
>
>
> `One of our clusters sometimes couldn't allocate blocks from any DNs.
> BlockPlacementPolicyDefault complains with the following messages for all DNs.
> {noformat}
> the node is too busy (load:x > y)
> {noformat}
> It turns out the {{HeartbeatManager}}'s {{nodesInService}} was computed
> incorrectly when admins decomm or recomm dead nodes. Here are two scenarios.
> * Decomm dead nodes. It turns out HDFS-7374 has fixed it; not sure if it is
> intentional. cc / [~zhz], [~andrew.wang], [~atm] Here is the sequence of
> event without HDFS-7374.
> ** Cluster has one live node. nodesInService == 1
> ** The node becomes dead. nodesInService == 0
> ** Decomm the node. nodesInService == -1
> * However, HDFS-7374 introduces another inconsistency when recomm is involved.
> ** Cluster has one live node. nodesInService == 1
> ** The node becomes dead. nodesInService == 0
> ** Decomm the node. nodesInService == 0
> ** Recomm the node. nodesInService == 1
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
[jira] [Updated] (HDFS-7725) Incorrect "nodes in service" metrics caused all writes to fail
[
https://issues.apache.org/jira/browse/HDFS-7725?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Ming Ma updated HDFS-7725:
--
Description:
One of our clusters sometimes couldn't allocate blocks from any DNs.
BlockPlacementPolicyDefault complains with the following messages for all DNs.
{noformat}
the node is too busy (load:x > y)
{noformat}
It turns out the {{HeartbeatManager}}'s {{nodesInService}} was computed
incorrectly when admins decomm or recomm dead nodes. Here are two scenarios.
* Decomm dead nodes. It turns out HDFS-7374 has fixed it; not sure if it is
intentional. cc / [~zhz], [~andrew.wang], [~atm] Here is the sequence of event
without HDFS-7374.
** Cluster has one live node. nodesInService == 1
** The node becomes dead. nodesInService == 0
** Decomm the node. nodesInService == -1
* However, HDFS-7374 introduces another inconsistency when recomm is involved.
** Cluster has one live node. nodesInService == 1
** The node becomes dead. nodesInService == 0
** Decomm the node. nodesInService == 0
** Recomm the node. nodesInService == 1
was:
One of our clusters sometimes couldn't allocate blocks from any DNs.
BlockPlacementPolicyDefault complains with the following messages for all DNs.
{noformat}
the node is too busy (load:x > y)
{noformat}
It turns out the {{HeartbeatManager}}'s {{nodesInService}} was computed
incorrectly when admins decomm or recomm dead nodes. Here are two scenarios.
* Decomm dead nodes. It turns out HDFS-7374 has fixed it; not sure if it is
intentional. Here is the sequence of event without HDFS-7374.
** Cluster has one live node. nodesInService == 1
** The node becomes dead. nodesInService == 0
** Decomm the node. nodesInService == -1
* However, HDFS-7374 introduces another inconsistency when recomm is involved.
** Cluster has one live node. nodesInService == 1
** The node becomes dead. nodesInService == 0
** Decomm the node. nodesInService == 0
** Recomm the node. nodesInService == 1
> Incorrect "nodes in service" metrics caused all writes to fail
> --
>
> Key: HDFS-7725
> URL: https://issues.apache.org/jira/browse/HDFS-7725
> Project: Hadoop HDFS
> Issue Type: Bug
>Reporter: Ming Ma
>Assignee: Ming Ma
> Attachments: HDFS-7725.patch
>
>
> One of our clusters sometimes couldn't allocate blocks from any DNs.
> BlockPlacementPolicyDefault complains with the following messages for all DNs.
> {noformat}
> the node is too busy (load:x > y)
> {noformat}
> It turns out the {{HeartbeatManager}}'s {{nodesInService}} was computed
> incorrectly when admins decomm or recomm dead nodes. Here are two scenarios.
> * Decomm dead nodes. It turns out HDFS-7374 has fixed it; not sure if it is
> intentional. cc / [~zhz], [~andrew.wang], [~atm] Here is the sequence of
> event without HDFS-7374.
> ** Cluster has one live node. nodesInService == 1
> ** The node becomes dead. nodesInService == 0
> ** Decomm the node. nodesInService == -1
> * However, HDFS-7374 introduces another inconsistency when recomm is involved.
> ** Cluster has one live node. nodesInService == 1
> ** The node becomes dead. nodesInService == 0
> ** Decomm the node. nodesInService == 0
> ** Recomm the node. nodesInService == 1
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
[jira] [Updated] (HDFS-7725) Incorrect "nodes in service" metrics caused all writes to fail
[
https://issues.apache.org/jira/browse/HDFS-7725?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Ming Ma updated HDFS-7725:
--
Attachment: HDFS-7725.patch
The patch makes sure the nodeInService count won't be updated when a dead node
is recommissioned.
> Incorrect "nodes in service" metrics caused all writes to fail
> --
>
> Key: HDFS-7725
> URL: https://issues.apache.org/jira/browse/HDFS-7725
> Project: Hadoop HDFS
> Issue Type: Bug
>Reporter: Ming Ma
> Attachments: HDFS-7725.patch
>
>
> One of our clusters sometimes couldn't allocate blocks from any DNs.
> BlockPlacementPolicyDefault complains with the following messages for all DNs.
> {noformat}
> the node is too busy (load:x > y)
> {noformat}
> It turns out the {{HeartbeatManager}}'s {{nodesInService}} was computed
> incorrectly when admins decomm or recomm dead nodes. Here are two scenarios.
> * Decomm dead nodes. It turns out HDFS-7374 has fixed it; not sure if it is
> intentional. Here is the sequence of event without HDFS-7374.
> ** Cluster has one live node. nodesInService == 1
> ** The node becomes dead. nodesInService == 0
> ** Decomm the node. nodesInService == -1
> * However, HDFS-7374 introduces another inconsistency when recomm is involved.
> ** Cluster has one live node. nodesInService == 1
> ** The node becomes dead. nodesInService == 0
> ** Decomm the node. nodesInService == 0
> ** Recomm the node. nodesInService == 1
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
[jira] [Updated] (HDFS-7725) Incorrect "nodes in service" metrics caused all writes to fail
[
https://issues.apache.org/jira/browse/HDFS-7725?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Ming Ma updated HDFS-7725:
--
Assignee: Ming Ma
Status: Patch Available (was: Open)
> Incorrect "nodes in service" metrics caused all writes to fail
> --
>
> Key: HDFS-7725
> URL: https://issues.apache.org/jira/browse/HDFS-7725
> Project: Hadoop HDFS
> Issue Type: Bug
>Reporter: Ming Ma
>Assignee: Ming Ma
> Attachments: HDFS-7725.patch
>
>
> One of our clusters sometimes couldn't allocate blocks from any DNs.
> BlockPlacementPolicyDefault complains with the following messages for all DNs.
> {noformat}
> the node is too busy (load:x > y)
> {noformat}
> It turns out the {{HeartbeatManager}}'s {{nodesInService}} was computed
> incorrectly when admins decomm or recomm dead nodes. Here are two scenarios.
> * Decomm dead nodes. It turns out HDFS-7374 has fixed it; not sure if it is
> intentional. Here is the sequence of event without HDFS-7374.
> ** Cluster has one live node. nodesInService == 1
> ** The node becomes dead. nodesInService == 0
> ** Decomm the node. nodesInService == -1
> * However, HDFS-7374 introduces another inconsistency when recomm is involved.
> ** Cluster has one live node. nodesInService == 1
> ** The node becomes dead. nodesInService == 0
> ** Decomm the node. nodesInService == 0
> ** Recomm the node. nodesInService == 1
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
