To be included in v0.9.3

Data replication (client side, synchronous)

This MR adds support for data replication using one environment variable:

`LIBGKFS_NUM_REPL=<num repl>`
The number of replicas should go from 0 to the number of servers-1. The replicas are guided by the 
client, so it reduces write performance but we mantain the same level of consistency. 
On the other hand, it may increase read performance on some corner scenearios. 
Metadata replication is also implemented
The replication environment variable can be set up for each client, independently.

If a server is down, the data will be read from another replica. The metadata management is also done from another replica.

The replication is done in a synchronous way. A new function forward_write is used to sent to the different replicas. The reads are distributed, but this shouldn't produce an performance improvement as the distribution is similar to the original. 

In the case of the write, the original is sent to the target servers, and then the replicas are processed. This is done to avoid issues if a server, that should host a replica, is not available. 

In order to process the replicas a new method to check that a chunk needs to be processed inside a server is included, a bitset of 1024 is sent (coded in base-64 in a string). This represents 1024-chunks per write-read operation. If that is exceeded the normal hash check per chunk is done in the server. Exceeding this value, will disable the replica capabilities and produce unknown behaviours.

This can be potentially increased.

Finally, most of the operations are replica-aware, but some of them are missing yet. i.e., dirent.

See merge request !166

Ramon Nou authored 1 year ago and Marc Vef committed 1 year ago

Added NUM_REPL env variable. (0 no replicas)

NUM_REPL  num replicas  (Replicas < servers)

Remove and truncate

Metadata replication

Metadata replication - Reattempt on stat

minimal compilation issues (c++20), srand for repl

Bitset

Bit set proposal (WIP)

Read - Write with bitset (<1024 chunks)

Changed bitset to vector

Added get_fs_config reattempt

Some more resilience on create

Added Replica_Check on write (disabled)

Added helper vector-bitset functions

Release 0.9.2


See merge request !181

Resolve "Client logging: Add per process log option"

Closes #278

Closes #278

See merge request !179

create logs in separate files for each process of the client.

Resolve "Support mtime"

Closes #225

Closes #225

See merge request !178

Resolve "du doesn't show size"

Closes #261

Closes #261

See merge request !177

calculates blocksize by dividing size / 512 as it is legacy in the linux kernel, only if renaming is deactivated

Resolve "Add GKFS_ prefix to CMake options"

Closes #129

Closes #129

See merge request !173

Resolve "Remove Parallax Testing"

This MR updates testing for the new docker CI environment:
- Removes Parallax Testing to make faster tests (we will activate again if needed, or changes are applied)
- Adds cstdint for uint types (gcc 10..)
- Updates python sh library/package to a newer version. This solves and produces some changes.
- Solved a bug in sfind

Closes #268

Closes #268

See merge request !169

Change all env PosixPath to str

restore rocksdb

Removed extra debug line

Added max internal fds

Reduce OPEN_FDS

Missing option

Resolve "Support Spack and others"

### Usage information

Download Spack and setup environment:
```bash
git clone -c feature.manyFiles=true https://github.com/spack/spack.git
. spack/share/spack/setup-env.sh
```
Add GekkoFS Spack repository to Spack:
```bash
spack repo add gekkofs/scripts/spack
```
Check that Spack can find GekkoFS:
```
spack info gekkofs
```
Install GekkoFS (and run optional tests). Check `spack info gekkofs` for available option and versions:
```bash
spack install gekkofs
# for installing tests dependencies and running tests
spack install -v --test=root gekkofs +tests
```
Load GekkoFS into environment:
```
spack load gekkofs
```
If you want to use the latest developer branch of GekkoFS:
```
spack install gekkofs@latest
```
The default is using version 0.9.1 the last stable release.

### TODO

- [x] Base Spack functionality, versions, and configuration support
- [x] Documentation
- [x] Advanced functionality, more detailed configuration support, e.g., Parallax and Prometheus
- [x] More easy way to get path to client library
- [x] Add GekkoFS client wrapper for `LD_PRELOAD`
- [ ] Add final version to main Spack repository if possible. (Not possible right now as it not clear how 3rd party libraries should be treated.





















Closes #58

Closes #58

See merge request !137