Mirroring splitter meta data
A method, system and computer program product for data replication comprising receiving an IO at a first storage processor (SP), sending metadata corresponding to the IO to a second SP, receiving an acknowledgement from the second SP indicating the second SP received the metadata and send the IO down an IO stack.
1. A system for data replication, the system comprising:
a storage system, the storage system comprising a storage medium, a first storage processor (SP) having a first splitter, a second SP having a second splitter; and
computer-executable logic operating in memory, wherein the computer-executable program logic is configured for execution of:
receiving an IO at the first SP;
sending metadata corresponding to the IO to the second SP;
receiving an acknowledgement from the second SP indicating the second SP received the metadata; and
send the IO down an IO stack of the storage system.
2. The system of claim 1 further comprising a data protection appliance (DPA) and wherein the computer executable logic is further configured for execution of:
sending the IO to the DPA.
3. The system of claim 2 wherein the computer executable logic is further configured for execution of:
sending, from the first SP, a flush marker to the second SP;
putting the flush marker into the metadata list at the second SP;
confirming the DPA write the meta data up to the flush marker; and
send the a flush metadata end command to the second SP.
4. The system of claim 2 wherein the computer executable logic is further configured for execution of:
determining that a splitter is down;
stopping replication;
reading metadata in the splitter that has not failed; and
re-syncing data to the DPA based on the metadata.
5. The system of claim 2 further comprising a second DPA and wherein the computer executable logic is further configured for execution of:
determining that a DPA is down;
stopping replication;
reading metadata in the splitters; and
re-syncing data to the second DPA based on the metadata.
6. The system of claim 3 further comprising a second DPA and wherein the computer executable logic is further configured for execution of:
freeing, by the splitter at the second storage processor, the metadata before the flush marker.
7. A computer program product for use in replication comprising:
a non-transitory computer readable medium encoded with computer executable program code for data replication, the code configured to enable the execution of:
receiving an IO at a first storage processor (SP) of a storage system, wherein the storage system has a second SP;
sending metadata corresponding to the IO to the second SP;
receiving an acknowledgement from the second SP indicating the second SP received the metadata; and
send the IO down an IO stack of the storage system.
8. The program product of claim 7 wherein the executable program code is further configured for execution of sending the IO to a DPA.
9. The program product of claim 8 wherein the executable program code is further configured for execution of:
sending, from the first SP, a flush marker to the second SP;
putting the flush marker into the metadata list at the second SP;
confirming the DPA write the meta data up to the flush marker; and
send the a flush metadata end command to the second SP.
10. The program product of claim 8 wherein the executable program code is further configured for execution of:
determining that a splitter is down;
stopping replication;
reading metadata in the splitter that has not failed; and
re-syncing data to the DPA based on the metadata.
11. The program product of claim 8 wherein the executable program code is further configured for execution of:
determining that a DPA is down;
stopping replication;
reading metadata in the splitters; and
re-syncing data to the second DPA based on the metadata.
12. The program product of claim 9 wherein the executable program code is further configured for execution of:
freeing, by the splitter at the second storage processor, the metadata before the flush marker.
13. A computer implemented method for data replication, the method comprising:
receiving an IO at a first storage processor (SP) of a storage system, wherein the storage system has a second SP;
sending metadata corresponding to the IO to the second SP;
receiving an acknowledgement from the second SP indicating the second SP received the metadata; and
send the IO down an IO stack of the storage system.
14. The computer implemented method of claim 13 wherein the method further comprising:
sending the IO to a DPA.
15. The computer implemented method of claim 14 wherein the method further comprising:
sending, from the first SP, a flush marker to the second SP;
putting the flush marker into the metadata list at the second SP;
confirming the DPA write the meta data up to the flush marker; and
send the a flush metadata end command to the second SP.
16. The computer implemented method of claim 14 wherein the method further comprising:
determining that a splitter is down;
stopping replication;
reading metadata in the splitter that has not failed; and
re-syncing data to the DPA based on the metadata.
17. The computer implemented method of claim 14 wherein the method further comprising:
determining that a DPA is down;
stopping replication;
reading metadata in the splitters; and
re-syncing data to the second DPA based on the metadata.
18. The computer implemented method of claim 14 wherein the method further comprising:
freeing by the splitter at the second storage processor, the metadata before the flush marker.