Re: [Gluster-devel] Question on choosing source of replica to heal with AFR
On 02/28/2014 01:09 PM, Zhang Huan wrote: On Fri, Feb 28, 2014 at 12:07 PM, Ravishankar N ravishan...@redhat.com mailto:ravishan...@redhat.comwrote: On 02/28/2014 07:28 AM, Zhang Huan wrote: Hello Ravi, Thanks for your reply. Sorry that I have a typo in my mail. It should by underlying corruption instead of underlying correction. I guess the logic of eliminating zero byte files from all innocent nodes is working for preventing underlying corruption to propagate to other brick. Asked in another way, if the underlying brick finds some file is corrupted, anything it could do to tell glusterfs to fix it? Hi Zhang, If all nodes are innocent (from AFR's point of view) ,then AFR cannot use the changelog attributes to determine which is source. In this case, the safest bet is to mark all zero byte files as sink, so that we don't end up healing in the wrong direction. Like I said earlier, AFR can only use the changelog attributes (xattrs) to determine the source/sinks. It cannot detect underlying on disk file system corruptions outside the scope of the xattrs. If you are sure that a particular brick is the right source despite the xattrs saying otherwise, you can manually change the attributes of the file on all bricks so that AFR now sees that brick as the source and heals in the expected direction. -Ravi Hello Ravi, IMO, changing the attributes might be dangerous, since concurrent access with glusterfs is introduced. Not sure if glusterfs has already provided some mechanism for this. You are right Zhang. My assumption was that the file wouldn't be modified from the mount point while you are modifying the xattrs at the bricks. My suggestion is to eliminate the zero-byte file from heal source even if is marked as a source. If the underlying filesystem finds some corruption (by scrubbing daemon after checking data checksum), it could truncate it to 0 and let glusterfs to do the healing job. If there is underlying FS corruption and we need to make gluster aware of it, then something like bit rot detection would be the way to go. You can find more information about some work in progress on the gluster website/ mailing list archives: http://www.gluster.org/community/documentation/index.php/Arch/BitRot_Detection http://lists.nongnu.org/archive/html/gluster-devel/2014-01/msg00209.html https://lists.gnu.org/archive/html/gluster-devel/2014-01/msg6.html -Ravi Here is several cases of analysis in my mind. 1. If this corrupted file is marked as the only source, then there is no correct replica in the filesystem (actually all are fools), just pick any one as the source to heal is OK; 2. If the corrupted file is one of the potential sources, eliminate this one should keep healing in the right direction without further corrupting other correct replicas. 3. If the corrupted file is not marked as a source, some other replica will be chosen as a source and this file will be overwritten with correct data. 4. If there is no one is marked as clean by attribute, it is quite unlikely this file is chosen as a source as its size is 0. Even it is chosen as a source, there is no further corruption of file content after heal. Zhang Huan ___ Gluster-devel mailing list Gluster-devel@nongnu.org https://lists.nongnu.org/mailman/listinfo/gluster-devel
Re: [Gluster-devel] Question on choosing source of replica to heal with AFR
On 02/26/2014 07:42 PM, Zhang Huan wrote: Hello guys, Anyone know about my question? Zhang Huan On Sun, Feb 23, 2014 at 11:28 AM, Zhang Huan zhh...@gmail.com mailto:zhh...@gmail.com wrote: Hello all, While reading codes about how to choose healing source, there is one thing that confuse me. Say we have 3 replica, and 2 of them are OK and the left one is outdated due to temporary IO failure. For some reason, one of the 2 correct replica is truncated to 0 due to some underlying correction. Will glusterfs kick the 0 size file out? or still consider it a correct one and may corrupt the left correct replica by healing? Out of the two correct replicas, gluster will pick the first healthy replica brick as source [see afr_sh_select_source()]. If that brick is truncated at the back-end due to 'underlying correction' (not sure what that means), then yes I'm afraid it will still be considered as correct source and you would get zero byte file in other 2 bricks because of the healing. In function afr_mark_sources(), it kicks 0 size file out when all nodes are innocent. Even when all nodes are fools, the file with largest size will be chosen as source. When it comes to the case that there is wise nodes, it won't further check file size. Considering different file size of replicate will trigger healing to work, I am wondering if there is any reason behind the code? The changelog extended attributes are marked by AFR based on the result of whether the file operation succeeded or not on each of the replica. It uses those attributes to determine the source/sink. Direct modification of the file at the brick will invalidate any meaning that the changelog holds. Thanks, Ravi Thanks. Zhang Huan ___ Gluster-devel mailing list Gluster-devel@nongnu.org https://lists.nongnu.org/mailman/listinfo/gluster-devel ___ Gluster-devel mailing list Gluster-devel@nongnu.org https://lists.nongnu.org/mailman/listinfo/gluster-devel
Re: [Gluster-devel] Question on choosing source of replica to heal with AFR
Hello Ravi, Thanks for your reply. Sorry that I have a typo in my mail. It should by underlying corruption instead of underlying correction. I guess the logic of eliminating zero byte files from all innocent nodes is working for preventing underlying corruption to propagate to other brick. Asked in another way, if the underlying brick finds some file is corrupted, anything it could do to tell glusterfs to fix it? Zhang Huan On Thu, Feb 27, 2014 at 7:50 PM, Ravishankar N ravishan...@redhat.comwrote: On 02/26/2014 07:42 PM, Zhang Huan wrote: Hello guys, Anyone know about my question? Zhang Huan On Sun, Feb 23, 2014 at 11:28 AM, Zhang Huan zhh...@gmail.com wrote: Hello all, While reading codes about how to choose healing source, there is one thing that confuse me. Say we have 3 replica, and 2 of them are OK and the left one is outdated due to temporary IO failure. For some reason, one of the 2 correct replica is truncated to 0 due to some underlying correction. Will glusterfs kick the 0 size file out? or still consider it a correct one and may corrupt the left correct replica by healing? Out of the two correct replicas, gluster will pick the first healthy replica brick as source [see afr_sh_select_source()]. If that brick is truncated at the back-end due to 'underlying correction' (not sure what that means), then yes I'm afraid it will still be considered as correct source and you would get zero byte file in other 2 bricks because of the healing. In function afr_mark_sources(), it kicks 0 size file out when all nodes are innocent. Even when all nodes are fools, the file with largest size will be chosen as source. When it comes to the case that there is wise nodes, it won't further check file size. Considering different file size of replicate will trigger healing to work, I am wondering if there is any reason behind the code? The changelog extended attributes are marked by AFR based on the result of whether the file operation succeeded or not on each of the replica. It uses those attributes to determine the source/sink. Direct modification of the file at the brick will invalidate any meaning that the changelog holds. Thanks, Ravi Thanks. Zhang Huan ___ Gluster-devel mailing listGluster-devel@nongnu.orghttps://lists.nongnu.org/mailman/listinfo/gluster-devel ___ Gluster-devel mailing list Gluster-devel@nongnu.org https://lists.nongnu.org/mailman/listinfo/gluster-devel
Re: [Gluster-devel] Question on choosing source of replica to heal with AFR
On 02/28/2014 07:28 AM, Zhang Huan wrote: Hello Ravi, Thanks for your reply. Sorry that I have a typo in my mail. It should by underlying corruption instead of underlying correction. I guess the logic of eliminating zero byte files from all innocent nodes is working for preventing underlying corruption to propagate to other brick. Asked in another way, if the underlying brick finds some file is corrupted, anything it could do to tell glusterfs to fix it? Hi Zhang, If all nodes are innocent (from AFR's point of view) ,then AFR cannot use the changelog attributes to determine which is source. In this case, the safest bet is to mark all zero byte files as sink, so that we don't end up healing in the wrong direction. Like I said earlier, AFR can only use the changelog attributes (xattrs) to determine the source/sinks. It cannot detect underlying on disk file system corruptions outside the scope of the xattrs. If you are sure that a particular brick is the right source despite the xattrs saying otherwise, you can manually change the attributes of the file on all bricks so that AFR now sees that brick as the source and heals in the expected direction. -Ravi Zhang Huan On Thu, Feb 27, 2014 at 7:50 PM, Ravishankar N ravishan...@redhat.com mailto:ravishan...@redhat.com wrote: On 02/26/2014 07:42 PM, Zhang Huan wrote: Hello guys, Anyone know about my question? Zhang Huan On Sun, Feb 23, 2014 at 11:28 AM, Zhang Huan zhh...@gmail.com mailto:zhh...@gmail.com wrote: Hello all, While reading codes about how to choose healing source, there is one thing that confuse me. Say we have 3 replica, and 2 of them are OK and the left one is outdated due to temporary IO failure. For some reason, one of the 2 correct replica is truncated to 0 due to some underlying correction. Will glusterfs kick the 0 size file out? or still consider it a correct one and may corrupt the left correct replica by healing? Out of the two correct replicas, gluster will pick the first healthy replica brick as source [see afr_sh_select_source()]. If that brick is truncated at the back-end due to 'underlying correction' (not sure what that means), then yes I'm afraid it will still be considered as correct source and you would get zero byte file in other 2 bricks because of the healing. In function afr_mark_sources(), it kicks 0 size file out when all nodes are innocent. Even when all nodes are fools, the file with largest size will be chosen as source. When it comes to the case that there is wise nodes, it won't further check file size. Considering different file size of replicate will trigger healing to work, I am wondering if there is any reason behind the code? The changelog extended attributes are marked by AFR based on the result of whether the file operation succeeded or not on each of the replica. It uses those attributes to determine the source/sink. Direct modification of the file at the brick will invalidate any meaning that the changelog holds. Thanks, Ravi Thanks. Zhang Huan ___ Gluster-devel mailing list Gluster-devel@nongnu.org mailto:Gluster-devel@nongnu.org https://lists.nongnu.org/mailman/listinfo/gluster-devel ___ Gluster-devel mailing list Gluster-devel@nongnu.org https://lists.nongnu.org/mailman/listinfo/gluster-devel
Re: [Gluster-devel] Question on choosing source of replica to heal with AFR
On Fri, Feb 28, 2014 at 12:07 PM, Ravishankar N ravishan...@redhat.comwrote: On 02/28/2014 07:28 AM, Zhang Huan wrote: Hello Ravi, Thanks for your reply. Sorry that I have a typo in my mail. It should by underlying corruption instead of underlying correction. I guess the logic of eliminating zero byte files from all innocent nodes is working for preventing underlying corruption to propagate to other brick. Asked in another way, if the underlying brick finds some file is corrupted, anything it could do to tell glusterfs to fix it? Hi Zhang, If all nodes are innocent (from AFR's point of view) ,then AFR cannot use the changelog attributes to determine which is source. In this case, the safest bet is to mark all zero byte files as sink, so that we don't end up healing in the wrong direction. Like I said earlier, AFR can only use the changelog attributes (xattrs) to determine the source/sinks. It cannot detect underlying on disk file system corruptions outside the scope of the xattrs. If you are sure that a particular brick is the right source despite the xattrs saying otherwise, you can manually change the attributes of the file on all bricks so that AFR now sees that brick as the source and heals in the expected direction. -Ravi Hello Ravi, IMO, changing the attributes might be dangerous, since concurrent access with glusterfs is introduced. Not sure if glusterfs has already provided some mechanism for this. My suggestion is to eliminate the zero-byte file from heal source even if is marked as a source. If the underlying filesystem finds some corruption (by scrubbing daemon after checking data checksum), it could truncate it to 0 and let glusterfs to do the healing job. Here is several cases of analysis in my mind. 1. If this corrupted file is marked as the only source, then there is no correct replica in the filesystem (actually all are fools), just pick any one as the source to heal is OK; 2. If the corrupted file is one of the potential sources, eliminate this one should keep healing in the right direction without further corrupting other correct replicas. 3. If the corrupted file is not marked as a source, some other replica will be chosen as a source and this file will be overwritten with correct data. 4. If there is no one is marked as clean by attribute, it is quite unlikely this file is chosen as a source as its size is 0. Even it is chosen as a source, there is no further corruption of file content after heal. Zhang Huan ___ Gluster-devel mailing list Gluster-devel@nongnu.org https://lists.nongnu.org/mailman/listinfo/gluster-devel
Re: [Gluster-devel] Question on choosing source of replica to heal with AFR
Hello guys, Anyone know about my question? Zhang Huan On Sun, Feb 23, 2014 at 11:28 AM, Zhang Huan zhh...@gmail.com wrote: Hello all, While reading codes about how to choose healing source, there is one thing that confuse me. Say we have 3 replica, and 2 of them are OK and the left one is outdated due to temporary IO failure. For some reason, one of the 2 correct replica is truncated to 0 due to some underlying correction. Will glusterfs kick the 0 size file out? or still consider it a correct one and may corrupt the left correct replica by healing? In function afr_mark_sources(), it kicks 0 size file out when all nodes are innocent. Even when all nodes are fools, the file with largest size will be chosen as source. When it comes to the case that there is wise nodes, it won't further check file size. Considering different file size of replicate will trigger healing to work, I am wondering if there is any reason behind the code? Thanks. Zhang Huan ___ Gluster-devel mailing list Gluster-devel@nongnu.org https://lists.nongnu.org/mailman/listinfo/gluster-devel