Harihara Kadayam

Results 21 issues of Harihara Kadayam

These reset apis are specifically useful when histograms are collected from different way and then convert them to prometheus metric.

* AppendBlkAllocator to implement alloc_on_cache and alloc_on_disk * FixedBlkAllocator didn't implement pre-alloc during recovery, implemented that

Fetch Blocks RPC when sent to existing member, responds with the data from HomeStore layer itself. The RPC sent all the neccessary information, . At present homestore uses remote_blkid (which...

Replication
GC

2 parts on this request 1. As we stand with current design for replication, we need to disable automatic timer flush in logdev. This would mean we split the logdev...

Logstore

Each replication request is tracked as repl_req structure. In followers they are created upon leader sending data either over raft or data channel (whichever comes first). There is a possibility...

Replication

With the revised Device Manager, we support partial boot up, as in even if one of the drives are unreadable or some regions are unreadable, we can still recover other...

enhancement
Device/Data

At present crash recovery of root node split and merge doesn't work correctly, because of presence of another node pointing to root node not persisted. We need to fix that.

Index Service

@xiaoxichen reported this issue while testing end to end flow. Possibly because bits avail isn't calculated correctly.

Device/Data

We need a generic mechanism to simulate crash and subsequent recovery for all unit tests to run. Without this, we are relying on actual crashing and then restart with some...

Use flip points in IOManager to simulate read/write failure during startup on different device classes and ensure that we can recover services partially.

enhancement
Device/Data