[jira] [Updated] (HBASE-10566) cleanup rpcTimeout in the client
[ https://issues.apache.org/jira/browse/HBASE-10566?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Dave Latham updated HBASE-10566: Release Note: 3 new settings are now available to configure the socket in the HBase client: - connect timeout: "hbase.ipc.client.socket.timeout.connect" (milliseconds, default: 10 seconds) - read timeout: "hbase.ipc.client.socket.timeout.read" (milliseconds, default: 20 seconds) - write timeout: "hbase.ipc.client.socket.timeout.write" (milliseconds, default: 60 seconds) ipc.socket.timeout is not used anymore. The per operation timeout is still controled by hbase.rpc.timeout was: 3 new settings are now available to configure the socket in the HBase client: - connect timeout: "hbase.ipc.client.socket.timeout.connect" (milliseconds, default: 10 seconds) - write timeout: "hbase.ipc.client.socket.timeout.read" (milliseconds, default: 20 seconds) - read timeout: "hbase.ipc.client.socket.timeout.write" (milliseconds, default: 60 seconds) ipc.socket.timeout is not used anymore. The per operation timeout is still controled by hbase.rpc.timeout > cleanup rpcTimeout in the client > > > Key: HBASE-10566 > URL: https://issues.apache.org/jira/browse/HBASE-10566 > Project: HBase > Issue Type: Bug > Components: Client >Affects Versions: 0.99.0 >Reporter: Nicolas Liochon >Assignee: Nicolas Liochon > Fix For: 0.99.0 > > Attachments: 10566.sample.patch, 10566.v1.patch, 10566.v2.patch, > 10566.v3.patch > > > There are two issues: > 1) A confusion between the socket timeout and the call timeout > Socket timeouts should be minimal: a default like 20 seconds, that could be > lowered to single digits timeouts for some apps: if we can not write to the > socket in 10 second, we have an issue. This is different from the total > duration (send query + do query + receive query), that can be longer, as it > can include remotes calls on the server and so on. Today, we have a single > value, it does not allow us to have low socket read timeouts. > 2) The timeout can be different between the calls. Typically, if the total > time, retries included is 60 seconds but failed after 2 seconds, then the > remaining is 58s. HBase does this today, but by hacking with a thread local > storage variable. It's a hack (it should have been a parameter of the > methods, the TLS allowed to bypass all the layers. May be protobuf makes this > complicated, to be confirmed), but as well it does not really work, because > we can have multithreading issues (we use the updated rpc timeout of someone > else, or we create a new BlockingRpcChannelImplementation with a random > default timeout). > Ideally, we could send the call timeout to the server as well: it will be > able to dismiss alone the calls that it received but git stick in the request > queue or in the internal retries (on hdfs for example). > This will make the system more reactive to failure. > I think we can solve this now, especially after 10525. The main issue is to > something that fits well with protobuf... > Then it should be easy to have a pool of thread for writers and readers, w/o > a single thread per region server as today. -- This message was sent by Atlassian JIRA (v6.4.14#64029)
[jira] [Updated] (HBASE-10566) cleanup rpcTimeout in the client
[ https://issues.apache.org/jira/browse/HBASE-10566?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Nicolas Liochon updated HBASE-10566: Release Note: 3 new settings are now available to configure the socket in the HBase client: - connect timeout: hbase.ipc.client.socket.timeout.connect (milliseconds, default: 10 seconds) - write timeout: hbase.ipc.client.socket.timeout.read (milliseconds, default: 20 seconds) - read timeout: hbase.ipc.client.socket.timeout.write (milliseconds, default: 60 seconds) ipc.socket.timeout is not used anymore. The per operation timeout is still controled by hbase.rpc.timeout was: 3 settings are now available to configure the socket in the HBase client: - connect timeout: ipc.socket.timeout.connect (default: 10 seconds) - write timeout: ipc.socket.timeout.read (default: 20 seconds) - read timeout: ipc.socket.timeout.write (default: 60 seconds) The per operation timeout is still controled by hbase.rpc.timeout cleanup rpcTimeout in the client Key: HBASE-10566 URL: https://issues.apache.org/jira/browse/HBASE-10566 Project: HBase Issue Type: Bug Components: Client Affects Versions: 0.99.0 Reporter: Nicolas Liochon Assignee: Nicolas Liochon Fix For: 0.99.0 Attachments: 10566.sample.patch, 10566.v1.patch, 10566.v2.patch, 10566.v3.patch There are two issues: 1) A confusion between the socket timeout and the call timeout Socket timeouts should be minimal: a default like 20 seconds, that could be lowered to single digits timeouts for some apps: if we can not write to the socket in 10 second, we have an issue. This is different from the total duration (send query + do query + receive query), that can be longer, as it can include remotes calls on the server and so on. Today, we have a single value, it does not allow us to have low socket read timeouts. 2) The timeout can be different between the calls. Typically, if the total time, retries included is 60 seconds but failed after 2 seconds, then the remaining is 58s. HBase does this today, but by hacking with a thread local storage variable. It's a hack (it should have been a parameter of the methods, the TLS allowed to bypass all the layers. May be protobuf makes this complicated, to be confirmed), but as well it does not really work, because we can have multithreading issues (we use the updated rpc timeout of someone else, or we create a new BlockingRpcChannelImplementation with a random default timeout). Ideally, we could send the call timeout to the server as well: it will be able to dismiss alone the calls that it received but git stick in the request queue or in the internal retries (on hdfs for example). This will make the system more reactive to failure. I think we can solve this now, especially after 10525. The main issue is to something that fits well with protobuf... Then it should be easy to have a pool of thread for writers and readers, w/o a single thread per region server as today. -- This message was sent by Atlassian JIRA (v6.1.5#6160)
[jira] [Updated] (HBASE-10566) cleanup rpcTimeout in the client
[ https://issues.apache.org/jira/browse/HBASE-10566?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Nicolas Liochon updated HBASE-10566: Status: Open (was: Patch Available) cleanup rpcTimeout in the client Key: HBASE-10566 URL: https://issues.apache.org/jira/browse/HBASE-10566 Project: HBase Issue Type: Bug Components: Client Affects Versions: 0.99.0 Reporter: Nicolas Liochon Assignee: Nicolas Liochon Fix For: 0.99.0 Attachments: 10566.sample.patch, 10566.v1.patch, 10566.v2.patch, 10566.v3.patch There are two issues: 1) A confusion between the socket timeout and the call timeout Socket timeouts should be minimal: a default like 20 seconds, that could be lowered to single digits timeouts for some apps: if we can not write to the socket in 10 second, we have an issue. This is different from the total duration (send query + do query + receive query), that can be longer, as it can include remotes calls on the server and so on. Today, we have a single value, it does not allow us to have low socket read timeouts. 2) The timeout can be different between the calls. Typically, if the total time, retries included is 60 seconds but failed after 2 seconds, then the remaining is 58s. HBase does this today, but by hacking with a thread local storage variable. It's a hack (it should have been a parameter of the methods, the TLS allowed to bypass all the layers. May be protobuf makes this complicated, to be confirmed), but as well it does not really work, because we can have multithreading issues (we use the updated rpc timeout of someone else, or we create a new BlockingRpcChannelImplementation with a random default timeout). Ideally, we could send the call timeout to the server as well: it will be able to dismiss alone the calls that it received but git stick in the request queue or in the internal retries (on hdfs for example). This will make the system more reactive to failure. I think we can solve this now, especially after 10525. The main issue is to something that fits well with protobuf... Then it should be easy to have a pool of thread for writers and readers, w/o a single thread per region server as today. -- This message was sent by Atlassian JIRA (v6.1.5#6160)
[jira] [Updated] (HBASE-10566) cleanup rpcTimeout in the client
[ https://issues.apache.org/jira/browse/HBASE-10566?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Nicolas Liochon updated HBASE-10566: Attachment: 10566.v3.patch cleanup rpcTimeout in the client Key: HBASE-10566 URL: https://issues.apache.org/jira/browse/HBASE-10566 Project: HBase Issue Type: Bug Components: Client Affects Versions: 0.99.0 Reporter: Nicolas Liochon Assignee: Nicolas Liochon Fix For: 0.99.0 Attachments: 10566.sample.patch, 10566.v1.patch, 10566.v2.patch, 10566.v3.patch There are two issues: 1) A confusion between the socket timeout and the call timeout Socket timeouts should be minimal: a default like 20 seconds, that could be lowered to single digits timeouts for some apps: if we can not write to the socket in 10 second, we have an issue. This is different from the total duration (send query + do query + receive query), that can be longer, as it can include remotes calls on the server and so on. Today, we have a single value, it does not allow us to have low socket read timeouts. 2) The timeout can be different between the calls. Typically, if the total time, retries included is 60 seconds but failed after 2 seconds, then the remaining is 58s. HBase does this today, but by hacking with a thread local storage variable. It's a hack (it should have been a parameter of the methods, the TLS allowed to bypass all the layers. May be protobuf makes this complicated, to be confirmed), but as well it does not really work, because we can have multithreading issues (we use the updated rpc timeout of someone else, or we create a new BlockingRpcChannelImplementation with a random default timeout). Ideally, we could send the call timeout to the server as well: it will be able to dismiss alone the calls that it received but git stick in the request queue or in the internal retries (on hdfs for example). This will make the system more reactive to failure. I think we can solve this now, especially after 10525. The main issue is to something that fits well with protobuf... Then it should be easy to have a pool of thread for writers and readers, w/o a single thread per region server as today. -- This message was sent by Atlassian JIRA (v6.1.5#6160)
[jira] [Updated] (HBASE-10566) cleanup rpcTimeout in the client
[ https://issues.apache.org/jira/browse/HBASE-10566?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Nicolas Liochon updated HBASE-10566: Status: Patch Available (was: Open) cleanup rpcTimeout in the client Key: HBASE-10566 URL: https://issues.apache.org/jira/browse/HBASE-10566 Project: HBase Issue Type: Bug Components: Client Affects Versions: 0.99.0 Reporter: Nicolas Liochon Assignee: Nicolas Liochon Fix For: 0.99.0 Attachments: 10566.sample.patch, 10566.v1.patch, 10566.v2.patch, 10566.v3.patch There are two issues: 1) A confusion between the socket timeout and the call timeout Socket timeouts should be minimal: a default like 20 seconds, that could be lowered to single digits timeouts for some apps: if we can not write to the socket in 10 second, we have an issue. This is different from the total duration (send query + do query + receive query), that can be longer, as it can include remotes calls on the server and so on. Today, we have a single value, it does not allow us to have low socket read timeouts. 2) The timeout can be different between the calls. Typically, if the total time, retries included is 60 seconds but failed after 2 seconds, then the remaining is 58s. HBase does this today, but by hacking with a thread local storage variable. It's a hack (it should have been a parameter of the methods, the TLS allowed to bypass all the layers. May be protobuf makes this complicated, to be confirmed), but as well it does not really work, because we can have multithreading issues (we use the updated rpc timeout of someone else, or we create a new BlockingRpcChannelImplementation with a random default timeout). Ideally, we could send the call timeout to the server as well: it will be able to dismiss alone the calls that it received but git stick in the request queue or in the internal retries (on hdfs for example). This will make the system more reactive to failure. I think we can solve this now, especially after 10525. The main issue is to something that fits well with protobuf... Then it should be easy to have a pool of thread for writers and readers, w/o a single thread per region server as today. -- This message was sent by Atlassian JIRA (v6.1.5#6160)
[jira] [Updated] (HBASE-10566) cleanup rpcTimeout in the client
[ https://issues.apache.org/jira/browse/HBASE-10566?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Nicolas Liochon updated HBASE-10566: Resolution: Fixed Hadoop Flags: Reviewed Status: Resolved (was: Patch Available) Committed to trunk, thanks for the review, Nick Stack. Release notes and derivate jiras are coming. cleanup rpcTimeout in the client Key: HBASE-10566 URL: https://issues.apache.org/jira/browse/HBASE-10566 Project: HBase Issue Type: Bug Components: Client Affects Versions: 0.99.0 Reporter: Nicolas Liochon Assignee: Nicolas Liochon Fix For: 0.99.0 Attachments: 10566.sample.patch, 10566.v1.patch, 10566.v2.patch, 10566.v3.patch There are two issues: 1) A confusion between the socket timeout and the call timeout Socket timeouts should be minimal: a default like 20 seconds, that could be lowered to single digits timeouts for some apps: if we can not write to the socket in 10 second, we have an issue. This is different from the total duration (send query + do query + receive query), that can be longer, as it can include remotes calls on the server and so on. Today, we have a single value, it does not allow us to have low socket read timeouts. 2) The timeout can be different between the calls. Typically, if the total time, retries included is 60 seconds but failed after 2 seconds, then the remaining is 58s. HBase does this today, but by hacking with a thread local storage variable. It's a hack (it should have been a parameter of the methods, the TLS allowed to bypass all the layers. May be protobuf makes this complicated, to be confirmed), but as well it does not really work, because we can have multithreading issues (we use the updated rpc timeout of someone else, or we create a new BlockingRpcChannelImplementation with a random default timeout). Ideally, we could send the call timeout to the server as well: it will be able to dismiss alone the calls that it received but git stick in the request queue or in the internal retries (on hdfs for example). This will make the system more reactive to failure. I think we can solve this now, especially after 10525. The main issue is to something that fits well with protobuf... Then it should be easy to have a pool of thread for writers and readers, w/o a single thread per region server as today. -- This message was sent by Atlassian JIRA (v6.1.5#6160)
[jira] [Updated] (HBASE-10566) cleanup rpcTimeout in the client
[ https://issues.apache.org/jira/browse/HBASE-10566?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Nicolas Liochon updated HBASE-10566: Release Note: 3 settings are now available to configure the socket in the HBase client: - connect timeout: ipc.socket.timeout.connect (default: 10 seconds) - write timeout: ipc.socket.timeout.read (default: 20 seconds) - read timeout: ipc.socket.timeout.write (default: 60 seconds) The per operation timeout is still controled by hbase.rpc.timeout cleanup rpcTimeout in the client Key: HBASE-10566 URL: https://issues.apache.org/jira/browse/HBASE-10566 Project: HBase Issue Type: Bug Components: Client Affects Versions: 0.99.0 Reporter: Nicolas Liochon Assignee: Nicolas Liochon Fix For: 0.99.0 Attachments: 10566.sample.patch, 10566.v1.patch, 10566.v2.patch, 10566.v3.patch There are two issues: 1) A confusion between the socket timeout and the call timeout Socket timeouts should be minimal: a default like 20 seconds, that could be lowered to single digits timeouts for some apps: if we can not write to the socket in 10 second, we have an issue. This is different from the total duration (send query + do query + receive query), that can be longer, as it can include remotes calls on the server and so on. Today, we have a single value, it does not allow us to have low socket read timeouts. 2) The timeout can be different between the calls. Typically, if the total time, retries included is 60 seconds but failed after 2 seconds, then the remaining is 58s. HBase does this today, but by hacking with a thread local storage variable. It's a hack (it should have been a parameter of the methods, the TLS allowed to bypass all the layers. May be protobuf makes this complicated, to be confirmed), but as well it does not really work, because we can have multithreading issues (we use the updated rpc timeout of someone else, or we create a new BlockingRpcChannelImplementation with a random default timeout). Ideally, we could send the call timeout to the server as well: it will be able to dismiss alone the calls that it received but git stick in the request queue or in the internal retries (on hdfs for example). This will make the system more reactive to failure. I think we can solve this now, especially after 10525. The main issue is to something that fits well with protobuf... Then it should be easy to have a pool of thread for writers and readers, w/o a single thread per region server as today. -- This message was sent by Atlassian JIRA (v6.1.5#6160)
[jira] [Updated] (HBASE-10566) cleanup rpcTimeout in the client
[ https://issues.apache.org/jira/browse/HBASE-10566?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Nicolas Liochon updated HBASE-10566: Status: Patch Available (was: Open) cleanup rpcTimeout in the client Key: HBASE-10566 URL: https://issues.apache.org/jira/browse/HBASE-10566 Project: HBase Issue Type: Bug Components: Client Affects Versions: 0.99.0 Reporter: Nicolas Liochon Assignee: Nicolas Liochon Fix For: 0.99.0 Attachments: 10566.sample.patch, 10566.v1.patch There are two issues: 1) A confusion between the socket timeout and the call timeout Socket timeouts should be minimal: a default like 20 seconds, that could be lowered to single digits timeouts for some apps: if we can not write to the socket in 10 second, we have an issue. This is different from the total duration (send query + do query + receive query), that can be longer, as it can include remotes calls on the server and so on. Today, we have a single value, it does not allow us to have low socket read timeouts. 2) The timeout can be different between the calls. Typically, if the total time, retries included is 60 seconds but failed after 2 seconds, then the remaining is 58s. HBase does this today, but by hacking with a thread local storage variable. It's a hack (it should have been a parameter of the methods, the TLS allowed to bypass all the layers. May be protobuf makes this complicated, to be confirmed), but as well it does not really work, because we can have multithreading issues (we use the updated rpc timeout of someone else, or we create a new BlockingRpcChannelImplementation with a random default timeout). Ideally, we could send the call timeout to the server as well: it will be able to dismiss alone the calls that it received but git stick in the request queue or in the internal retries (on hdfs for example). This will make the system more reactive to failure. I think we can solve this now, especially after 10525. The main issue is to something that fits well with protobuf... Then it should be easy to have a pool of thread for writers and readers, w/o a single thread per region server as today. -- This message was sent by Atlassian JIRA (v6.1.5#6160)
[jira] [Updated] (HBASE-10566) cleanup rpcTimeout in the client
[ https://issues.apache.org/jira/browse/HBASE-10566?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Nicolas Liochon updated HBASE-10566: Attachment: 10566.v1.patch cleanup rpcTimeout in the client Key: HBASE-10566 URL: https://issues.apache.org/jira/browse/HBASE-10566 Project: HBase Issue Type: Bug Components: Client Affects Versions: 0.99.0 Reporter: Nicolas Liochon Assignee: Nicolas Liochon Fix For: 0.99.0 Attachments: 10566.sample.patch, 10566.v1.patch There are two issues: 1) A confusion between the socket timeout and the call timeout Socket timeouts should be minimal: a default like 20 seconds, that could be lowered to single digits timeouts for some apps: if we can not write to the socket in 10 second, we have an issue. This is different from the total duration (send query + do query + receive query), that can be longer, as it can include remotes calls on the server and so on. Today, we have a single value, it does not allow us to have low socket read timeouts. 2) The timeout can be different between the calls. Typically, if the total time, retries included is 60 seconds but failed after 2 seconds, then the remaining is 58s. HBase does this today, but by hacking with a thread local storage variable. It's a hack (it should have been a parameter of the methods, the TLS allowed to bypass all the layers. May be protobuf makes this complicated, to be confirmed), but as well it does not really work, because we can have multithreading issues (we use the updated rpc timeout of someone else, or we create a new BlockingRpcChannelImplementation with a random default timeout). Ideally, we could send the call timeout to the server as well: it will be able to dismiss alone the calls that it received but git stick in the request queue or in the internal retries (on hdfs for example). This will make the system more reactive to failure. I think we can solve this now, especially after 10525. The main issue is to something that fits well with protobuf... Then it should be easy to have a pool of thread for writers and readers, w/o a single thread per region server as today. -- This message was sent by Atlassian JIRA (v6.1.5#6160)
[jira] [Updated] (HBASE-10566) cleanup rpcTimeout in the client
[ https://issues.apache.org/jira/browse/HBASE-10566?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Nicolas Liochon updated HBASE-10566: Status: Open (was: Patch Available) cleanup rpcTimeout in the client Key: HBASE-10566 URL: https://issues.apache.org/jira/browse/HBASE-10566 Project: HBase Issue Type: Bug Components: Client Affects Versions: 0.99.0 Reporter: Nicolas Liochon Assignee: Nicolas Liochon Fix For: 0.99.0 Attachments: 10566.sample.patch, 10566.v1.patch, 10566.v2.patch There are two issues: 1) A confusion between the socket timeout and the call timeout Socket timeouts should be minimal: a default like 20 seconds, that could be lowered to single digits timeouts for some apps: if we can not write to the socket in 10 second, we have an issue. This is different from the total duration (send query + do query + receive query), that can be longer, as it can include remotes calls on the server and so on. Today, we have a single value, it does not allow us to have low socket read timeouts. 2) The timeout can be different between the calls. Typically, if the total time, retries included is 60 seconds but failed after 2 seconds, then the remaining is 58s. HBase does this today, but by hacking with a thread local storage variable. It's a hack (it should have been a parameter of the methods, the TLS allowed to bypass all the layers. May be protobuf makes this complicated, to be confirmed), but as well it does not really work, because we can have multithreading issues (we use the updated rpc timeout of someone else, or we create a new BlockingRpcChannelImplementation with a random default timeout). Ideally, we could send the call timeout to the server as well: it will be able to dismiss alone the calls that it received but git stick in the request queue or in the internal retries (on hdfs for example). This will make the system more reactive to failure. I think we can solve this now, especially after 10525. The main issue is to something that fits well with protobuf... Then it should be easy to have a pool of thread for writers and readers, w/o a single thread per region server as today. -- This message was sent by Atlassian JIRA (v6.1.5#6160)
[jira] [Updated] (HBASE-10566) cleanup rpcTimeout in the client
[ https://issues.apache.org/jira/browse/HBASE-10566?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Nicolas Liochon updated HBASE-10566: Status: Patch Available (was: Open) cleanup rpcTimeout in the client Key: HBASE-10566 URL: https://issues.apache.org/jira/browse/HBASE-10566 Project: HBase Issue Type: Bug Components: Client Affects Versions: 0.99.0 Reporter: Nicolas Liochon Assignee: Nicolas Liochon Fix For: 0.99.0 Attachments: 10566.sample.patch, 10566.v1.patch, 10566.v2.patch There are two issues: 1) A confusion between the socket timeout and the call timeout Socket timeouts should be minimal: a default like 20 seconds, that could be lowered to single digits timeouts for some apps: if we can not write to the socket in 10 second, we have an issue. This is different from the total duration (send query + do query + receive query), that can be longer, as it can include remotes calls on the server and so on. Today, we have a single value, it does not allow us to have low socket read timeouts. 2) The timeout can be different between the calls. Typically, if the total time, retries included is 60 seconds but failed after 2 seconds, then the remaining is 58s. HBase does this today, but by hacking with a thread local storage variable. It's a hack (it should have been a parameter of the methods, the TLS allowed to bypass all the layers. May be protobuf makes this complicated, to be confirmed), but as well it does not really work, because we can have multithreading issues (we use the updated rpc timeout of someone else, or we create a new BlockingRpcChannelImplementation with a random default timeout). Ideally, we could send the call timeout to the server as well: it will be able to dismiss alone the calls that it received but git stick in the request queue or in the internal retries (on hdfs for example). This will make the system more reactive to failure. I think we can solve this now, especially after 10525. The main issue is to something that fits well with protobuf... Then it should be easy to have a pool of thread for writers and readers, w/o a single thread per region server as today. -- This message was sent by Atlassian JIRA (v6.1.5#6160)
[jira] [Updated] (HBASE-10566) cleanup rpcTimeout in the client
[ https://issues.apache.org/jira/browse/HBASE-10566?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Nicolas Liochon updated HBASE-10566: Attachment: 10566.sample.patch cleanup rpcTimeout in the client Key: HBASE-10566 URL: https://issues.apache.org/jira/browse/HBASE-10566 Project: HBase Issue Type: Bug Components: Client Affects Versions: 0.99.0 Reporter: Nicolas Liochon Assignee: Nicolas Liochon Fix For: 0.99.0 Attachments: 10566.sample.patch There are two issues: 1) A confusion between the socket timeout and the call timeout Socket timeouts should be minimal: a default like 20 seconds, that could be lowered to single digits timeouts for some apps: if we can not write to the socket in 10 second, we have an issue. This is different from the total duration (send query + do query + receive query), that can be longer, as it can include remotes calls on the server and so on. Today, we have a single value, it does not allow us to have low socket read timeouts. 2) The timeout can be different between the calls. Typically, if the total time, retries included is 60 seconds but failed after 2 seconds, then the remaining is 58s. HBase does this today, but by hacking with a thread local storage variable. It's a hack (it should have been a parameter of the methods, the TLS allowed to bypass all the layers. May be protobuf makes this complicated, to be confirmed), but as well it does not really work, because we can have multithreading issues (we use the updated rpc timeout of someone else, or we create a new BlockingRpcChannelImplementation with a random default timeout). Ideally, we could send the call timeout to the server as well: it will be able to dismiss alone the calls that it received but git stick in the request queue or in the internal retries (on hdfs for example). This will make the system more reactive to failure. I think we can solve this now, especially after 10525. The main issue is to something that fits well with protobuf... Then it should be easy to have a pool of thread for writers and readers, w/o a single thread per region server as today. -- This message was sent by Atlassian JIRA (v6.1.5#6160)