[jira] [Commented] (CASSANDRA-6364) There should be different disk_failure_policies for data and commit volumes or commit volume failure should always cause node exit
[ https://issues.apache.org/jira/browse/CASSANDRA-6364?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=13897791#comment-13897791 ] Marcus Eriksson commented on CASSANDRA-6364: +1 on the 2.0-patch, committed There should be different disk_failure_policies for data and commit volumes or commit volume failure should always cause node exit -- Key: CASSANDRA-6364 URL: https://issues.apache.org/jira/browse/CASSANDRA-6364 Project: Cassandra Issue Type: Improvement Components: Core Environment: JBOD, single dedicated commit disk Reporter: J. Ryan Earl Assignee: Benedict Fix For: 2.0.6 Attachments: tmp-2.0.patch We're doing fault testing on a pre-production Cassandra cluster. One of the tests was to simulation failure of the commit volume/disk, which in our case is on a dedicated disk. We expected failure of the commit volume to be handled somehow, but what we found was that no action was taken by Cassandra when the commit volume fail. We simulated this simply by pulling the physical disk that backed the commit volume, which resulted in filesystem I/O errors on the mount point. What then happened was that the Cassandra Heap filled up to the point that it was spending 90% of its time doing garbage collection. No errors were logged in regards to the failed commit volume. Gossip on other nodes in the cluster eventually flagged the node as down. Gossip on the local node showed itself as up, and all other nodes as down. The most serious problem was that connections to the coordinator on this node became very slow due to the on-going GC, as I assume uncommitted writes piled up on the JVM heap. What we believe should have happened is that Cassandra should have caught the I/O error and exited with a useful log message, or otherwise done some sort of useful cleanup. Otherwise the node goes into a sort of Zombie state, spending most of its time in GC, and thus slowing down any transactions that happen to use the coordinator on said node. A limit on in-memory, unflushed writes before refusing requests may also work. Point being, something should be done to handle the commit volume dying as doing nothing results in affecting the entire cluster. I should note, we are using: disk_failure_policy: best_effort -- This message was sent by Atlassian JIRA (v6.1.5#6160)
[jira] [Commented] (CASSANDRA-6364) There should be different disk_failure_policies for data and commit volumes or commit volume failure should always cause node exit
[ https://issues.apache.org/jira/browse/CASSANDRA-6364?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=13889330#comment-13889330 ] Marcus Eriksson commented on CASSANDRA-6364: About the ignore case, lets hard code something for now - rate limit at one log error message per second perhaps? I don't think we should default to 'ignore' in Config.java - if someone does a minor upgrade they most likely wont check NEWS or update their config files to add the new parameter. The shipped config in cassandra.yaml looks wrong, should be commit_failure_policy, not disk_failure_policy I guess There should be different disk_failure_policies for data and commit volumes or commit volume failure should always cause node exit -- Key: CASSANDRA-6364 URL: https://issues.apache.org/jira/browse/CASSANDRA-6364 Project: Cassandra Issue Type: Improvement Components: Core Environment: JBOD, single dedicated commit disk Reporter: J. Ryan Earl Assignee: Benedict Fix For: 2.0.5 We're doing fault testing on a pre-production Cassandra cluster. One of the tests was to simulation failure of the commit volume/disk, which in our case is on a dedicated disk. We expected failure of the commit volume to be handled somehow, but what we found was that no action was taken by Cassandra when the commit volume fail. We simulated this simply by pulling the physical disk that backed the commit volume, which resulted in filesystem I/O errors on the mount point. What then happened was that the Cassandra Heap filled up to the point that it was spending 90% of its time doing garbage collection. No errors were logged in regards to the failed commit volume. Gossip on other nodes in the cluster eventually flagged the node as down. Gossip on the local node showed itself as up, and all other nodes as down. The most serious problem was that connections to the coordinator on this node became very slow due to the on-going GC, as I assume uncommitted writes piled up on the JVM heap. What we believe should have happened is that Cassandra should have caught the I/O error and exited with a useful log message, or otherwise done some sort of useful cleanup. Otherwise the node goes into a sort of Zombie state, spending most of its time in GC, and thus slowing down any transactions that happen to use the coordinator on said node. A limit on in-memory, unflushed writes before refusing requests may also work. Point being, something should be done to handle the commit volume dying as doing nothing results in affecting the entire cluster. I should note, we are using: disk_failure_policy: best_effort -- This message was sent by Atlassian JIRA (v6.1.5#6160)
[jira] [Commented] (CASSANDRA-6364) There should be different disk_failure_policies for data and commit volumes or commit volume failure should always cause node exit
[ https://issues.apache.org/jira/browse/CASSANDRA-6364?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=13889343#comment-13889343 ] Benedict commented on CASSANDRA-6364: - bq. I don't think we should default to 'ignore' in Config.java Well, I wasn't too sure about this. On the one hand switching the default to stop means we could over cautiously kill user's hosts unexpectedly, maybe resulting in interruption of service (especially, say, our users running on SAN, as much as it is strongly discouraged). Whereas switching to ignore means we may not be durable. Neither are great defaults, but both are better than before. I'm comfortable with both, so if you feel strongly it should be stop, I'll happily switch it. Perhaps I lean slightly in favour of it too, but it depends on if the user favours durability over availability, really, so there doesn't seem a single correct answer to me. Note that the default disk_failure_policy is also ignore, and the prior behaviour was closest to ignore, so introducing a default that results in a failing node is somewhat unprecedented for disk failure. bq. The shipped config in cassandra.yaml looks wrong, should be commit_failure_policy, not disk_failure_policy I guess Right, looks like I didn't update the first or last lines I copy-pasted. Thanks. bq. About the ignore case, lets hard code something for now - rate limit at one log error message per second perhaps? If we're just rate limiting the log messages, I'd say one per minute might be better. But I'm not sure having the threads spin trying to make progress is useful. The PCLES, for instance, will just start burning one core until it can successfully sync, assuming it doesn't actually have to wait each time to encounter the error. Tempted to have a 1s pause after an error during which we just sleep the erroring thread. Another issue that slightly concerns me is what happens if the CLES sync() starts failing, but the append and CLA doesn't. With ignore this could potentially result in us mapping in and allocating huge amounts of disk space, but not being able to sync or clear it. This might either result in lots of swapping, and/or us exceeding by a large margin our max log space goal. Since we never guarantee to keep to this I'm not sure how much of a problem it would be, but an error down to ACLs that stops us syncing one file might potentally end up eating up huge quantities of commit disk space. I'm tempted to have the CLA thread block once it hits twice its goal max space (or maybe introduce a second config parameter for a hard maximum). But I'm also tempted to leave these changes for the 2.1 branch, since it's a fairly specific failure case, and what we have is a big improvement over the current state of affairs. There should be different disk_failure_policies for data and commit volumes or commit volume failure should always cause node exit -- Key: CASSANDRA-6364 URL: https://issues.apache.org/jira/browse/CASSANDRA-6364 Project: Cassandra Issue Type: Improvement Components: Core Environment: JBOD, single dedicated commit disk Reporter: J. Ryan Earl Assignee: Benedict Fix For: 2.0.5 We're doing fault testing on a pre-production Cassandra cluster. One of the tests was to simulation failure of the commit volume/disk, which in our case is on a dedicated disk. We expected failure of the commit volume to be handled somehow, but what we found was that no action was taken by Cassandra when the commit volume fail. We simulated this simply by pulling the physical disk that backed the commit volume, which resulted in filesystem I/O errors on the mount point. What then happened was that the Cassandra Heap filled up to the point that it was spending 90% of its time doing garbage collection. No errors were logged in regards to the failed commit volume. Gossip on other nodes in the cluster eventually flagged the node as down. Gossip on the local node showed itself as up, and all other nodes as down. The most serious problem was that connections to the coordinator on this node became very slow due to the on-going GC, as I assume uncommitted writes piled up on the JVM heap. What we believe should have happened is that Cassandra should have caught the I/O error and exited with a useful log message, or otherwise done some sort of useful cleanup. Otherwise the node goes into a sort of Zombie state, spending most of its time in GC, and thus slowing down any transactions that happen to use the coordinator on said node. A limit on in-memory, unflushed writes before refusing requests may also work. Point being, something should
[jira] [Commented] (CASSANDRA-6364) There should be different disk_failure_policies for data and commit volumes or commit volume failure should always cause node exit
[ https://issues.apache.org/jira/browse/CASSANDRA-6364?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=13838786#comment-13838786 ] Benedict commented on CASSANDRA-6364: - How far do we want to go with this? Adding a simple exit on error is very straightforward, but in my experience you can have hang-style failures, so we should definitely have a separate thread checking the liveness of the CLSegmentManager and CLService. Probably a user-configurable not-alive time in the yaml should be used to mark the CL as non-responsive if either hasn't heartbeated in that time. Probably we don't want to immediately die on an error too, but simply not heartbeat and die if the error doesn't recover in some interval, so that anyone monitoring the error logs has time to correct the issue (let's say it's just out of space) before it dies. The bigger question is, do we want to do anything clever if we don't want to die? Should we start draining the mutation stage and just dropping the messages? If so, should we attempt to recover if the drive starts responding again after draining the mutation stage? There should be different disk_failure_policies for data and commit volumes or commit volume failure should always cause node exit -- Key: CASSANDRA-6364 URL: https://issues.apache.org/jira/browse/CASSANDRA-6364 Project: Cassandra Issue Type: Improvement Components: Core Environment: JBOD, single dedicated commit disk Reporter: J. Ryan Earl Assignee: Benedict Fix For: 2.0.4 We're doing fault testing on a pre-production Cassandra cluster. One of the tests was to simulation failure of the commit volume/disk, which in our case is on a dedicated disk. We expected failure of the commit volume to be handled somehow, but what we found was that no action was taken by Cassandra when the commit volume fail. We simulated this simply by pulling the physical disk that backed the commit volume, which resulted in filesystem I/O errors on the mount point. What then happened was that the Cassandra Heap filled up to the point that it was spending 90% of its time doing garbage collection. No errors were logged in regards to the failed commit volume. Gossip on other nodes in the cluster eventually flagged the node as down. Gossip on the local node showed itself as up, and all other nodes as down. The most serious problem was that connections to the coordinator on this node became very slow due to the on-going GC, as I assume uncommitted writes piled up on the JVM heap. What we believe should have happened is that Cassandra should have caught the I/O error and exited with a useful log message, or otherwise done some sort of useful cleanup. Otherwise the node goes into a sort of Zombie state, spending most of its time in GC, and thus slowing down any transactions that happen to use the coordinator on said node. A limit on in-memory, unflushed writes before refusing requests may also work. Point being, something should be done to handle the commit volume dying as doing nothing results in affecting the entire cluster. I should note, we are using: disk_failure_policy: best_effort -- This message was sent by Atlassian JIRA (v6.1#6144)
[jira] [Commented] (CASSANDRA-6364) There should be different disk_failure_policies for data and commit volumes or commit volume failure should always cause node exit
[ https://issues.apache.org/jira/browse/CASSANDRA-6364?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=13839383#comment-13839383 ] Jonathan Ellis commented on CASSANDRA-6364: --- Just make it die on IOError like the existing code. For now people can deal with hangs instead of erroring manually. There should be different disk_failure_policies for data and commit volumes or commit volume failure should always cause node exit -- Key: CASSANDRA-6364 URL: https://issues.apache.org/jira/browse/CASSANDRA-6364 Project: Cassandra Issue Type: Improvement Components: Core Environment: JBOD, single dedicated commit disk Reporter: J. Ryan Earl Assignee: Benedict Fix For: 2.0.4 We're doing fault testing on a pre-production Cassandra cluster. One of the tests was to simulation failure of the commit volume/disk, which in our case is on a dedicated disk. We expected failure of the commit volume to be handled somehow, but what we found was that no action was taken by Cassandra when the commit volume fail. We simulated this simply by pulling the physical disk that backed the commit volume, which resulted in filesystem I/O errors on the mount point. What then happened was that the Cassandra Heap filled up to the point that it was spending 90% of its time doing garbage collection. No errors were logged in regards to the failed commit volume. Gossip on other nodes in the cluster eventually flagged the node as down. Gossip on the local node showed itself as up, and all other nodes as down. The most serious problem was that connections to the coordinator on this node became very slow due to the on-going GC, as I assume uncommitted writes piled up on the JVM heap. What we believe should have happened is that Cassandra should have caught the I/O error and exited with a useful log message, or otherwise done some sort of useful cleanup. Otherwise the node goes into a sort of Zombie state, spending most of its time in GC, and thus slowing down any transactions that happen to use the coordinator on said node. A limit on in-memory, unflushed writes before refusing requests may also work. Point being, something should be done to handle the commit volume dying as doing nothing results in affecting the entire cluster. I should note, we are using: disk_failure_policy: best_effort -- This message was sent by Atlassian JIRA (v6.1#6144)
[jira] [Commented] (CASSANDRA-6364) There should be different disk_failure_policies for data and commit volumes or commit volume failure should always cause node exit
[ https://issues.apache.org/jira/browse/CASSANDRA-6364?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=13828064#comment-13828064 ] Aleksey Yeschenko commented on CASSANDRA-6364: -- I think I agree w/ stopping upon getting a write error on the commitlog drive. There should be different disk_failure_policies for data and commit volumes or commit volume failure should always cause node exit -- Key: CASSANDRA-6364 URL: https://issues.apache.org/jira/browse/CASSANDRA-6364 Project: Cassandra Issue Type: Improvement Environment: JBOD, single dedicated commit disk Reporter: J. Ryan Earl We're doing fault testing on a pre-production Cassandra cluster. One of the tests was to simulation failure of the commit volume/disk, which in our case is on a dedicated disk. We expected failure of the commit volume to be handled somehow, but what we found was that no action was taken by Cassandra when the commit volume fail. We simulated this simply by pulling the physical disk that backed the commit volume, which resulted in filesystem I/O errors on the mount point. What then happened was that the Cassandra Heap filled up to the point that it was spending 90% of its time doing garbage collection. No errors were logged in regards to the failed commit volume. Gossip on other nodes in the cluster eventually flagged the node as down. Gossip on the local node showed itself as up, and all other nodes as down. The most serious problem was that connections to the coordinator on this node became very slow due to the on-going GC, as I assume uncommitted writes piled up on the JVM heap. What we believe should have happened is that Cassandra should have caught the I/O error and exited with a useful log message, or otherwise done some sort of useful cleanup. Otherwise the node goes into a sort of Zombie state, spending most of its time in GC, and thus slowing down any transactions that happen to use the coordinator on said node. A limit on in-memory, unflushed writes before refusing requests may also work. Point being, something should be done to handle the commit volume dying as doing nothing results in affecting the entire cluster. I should note, we are using: disk_failure_policy: best_effort -- This message was sent by Atlassian JIRA (v6.1#6144)