audron/garage
1
0
Fork 0
forked from Deuxfleurs/garage
Code Issues Pull requests Projects Releases Wiki Activity

Add checks on replication_factor of layouts we use (fix #363, fix #364)

Browse source
This commit is contained in:
Alex 2022-09-13 16:22:23 +02:00
parent 38be811b1c
commit ab722cb40f
Signed by untrusted user: lx
GPG key ID: 0E496D15096376BE
2 changed files with 26 additions and 6 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

2
src/model/garage.rs
View file

@ -169,7 +169,7 @@ impl Garage {
background.clone(),
replication_mode.replication_factor(),
&config,
);
)?;
let data_rep_param = TableShardedReplication {
system: system.clone(),

30
src/rpc/system.rs
View file

@ -198,7 +198,7 @@ impl System {
background: Arc<BackgroundRunner>,
replication_factor: usize,
config: &Config,
) -> Arc<Self> {
) -> Result<Arc<Self>, Error> {
let node_key =
gen_node_key(&config.metadata_dir).expect("Unable to read or generate node ID");
info!(
@ -206,11 +206,21 @@ impl System {
hex::encode(&node_key.public_key()[..8])
);
let persist_cluster_layout = Persister::new(&config.metadata_dir, "cluster_layout");
let persist_cluster_layout: Persister<ClusterLayout> =
Persister::new(&config.metadata_dir, "cluster_layout");
let persist_peer_list = Persister::new(&config.metadata_dir, "peer_list");
let cluster_layout = match persist_cluster_layout.load() {
Ok(x) => x,
Ok(x) => {
if x.replication_factor != replication_factor {
return Err(Error::Message(format!(
"Prevous cluster layout has replication factor {}, which is different than the one specified in the config file ({}). The previous cluster layout can be purged, if you know what you are doing, simply by deleting the `cluster_layout` file in your metadata directory.",
x.replication_factor,
replication_factor
)));
}
x
}
Err(e) => {
info!(
"No valid previous cluster layout stored ({}), starting fresh.",
@ -303,7 +313,7 @@ impl System {
metadata_dir: config.metadata_dir.clone(),
});
sys.system_endpoint.set_handler(sys.clone());
sys
Ok(sys)
}
/// Perform bootstraping, starting the ping loop
@ -485,7 +495,7 @@ impl System {
let local_info = self.local_status.load();
if local_info.replication_factor < info.replication_factor {
error!("Some node have a higher replication factor ({}) than this one ({}). This is not supported and might lead to bugs",
error!("Some node have a higher replication factor ({}) than this one ({}). This is not supported and will lead to data corruption. Shutting down for safety.",
info.replication_factor,
local_info.replication_factor);
std::process::exit(1);
@ -513,6 +523,16 @@ impl System {
self: &Arc<Self>,
adv: &ClusterLayout,
) -> Result<SystemRpc, Error> {
if adv.replication_factor != self.replication_factor {
let msg = format!(
"Received a cluster layout from another node with replication factor {}, which is different from what we have in our configuration ({}). Discarding the cluster layout we received.",
adv.replication_factor,
self.replication_factor
);
error!("{}", msg);
return Err(Error::Message(msg));
}
let update_ring = self.update_ring.lock().await;
let mut layout: ClusterLayout = self.ring.borrow().layout.clone();