Closed rizcompeer closed 9 years ago
I found the cause of this. Cassandra is not suited for doing rapid deletes and updates with a multi region cluster where a delete may be propagated after some delay and end-up deleting updated data. This happened because we had a multi-region cluster with one node in Singapore. The correct way to do this is to drain all nodes after delete and then update.
For now, I do not want to delete any data. So just give me imei and dates which I should insert.
Missing Imei data is given below -
IMEI
DATE
865733021571229
2015-06-25
865733021571096
2015-06-19 to 2015-07-31
865733021569959
2015-06-24 to 2015-07-31
865733021571088
2015-06-26 to 2015-07-31
865733021571237
2015-06-28 to 2015-07-31
865733021562939
2015-06-28 to 2015-06-30, 2015-07-18
On Wed, Aug 19, 2015 at 4:45 PM, Abhishek Singh Bailoo < notifications@github.com> wrote:
I found the cause of this. Cassandra is not suited for doing rapid deletes and updates with a multi region cluster where a delete may be propagated after some delay and end-up deleting updated data. This happened because we had a multi-region cluster with one node in Singapore. The correct way to do this is to drain all nodes after delete and then update.
For now, I do not want to delete any data. So just give me imei and dates which I should insert.
— Reply to this email directly or view it on GitHub https://github.com/bailoo/itrack/issues/59#issuecomment-132550406.
Regards
Rizwan Ahmad Innovative Embedded Systems.Pvt.Ltd.
Done please check.
Excellent, master bailoo On 19 Aug 2015 16:45, "Abhishek Singh Bailoo" notifications@github.com wrote:
I found the cause of this. Cassandra is not suited for doing rapid deletes and updates with a multi region cluster where a delete may be propagated after some delay and end-up deleting updated data. This happened because we had a multi-region cluster with one node in Singapore. The correct way to do this is to drain all nodes after delete and then update.
For now, I do not want to delete any data. So just give me imei and dates which I should insert.
— Reply to this email directly or view it on GitHub https://github.com/bailoo/itrack/issues/59#issuecomment-132550406.
865733021571229
Few data of Jun,July of HDFC is missing, detail is given below -
865733021571237
865733021571096
865733021571088
865730321569959
865733021562939 (28,29,30 June)