Bug: NFS COPY post-commit checkpoint failure cause core dump

[ericyuanhui]

Ladybug version

v0.19.0

What operating system are you using?

ubuntu24

What happened?

Incident and cause
During a large COPY FROM, stopping the NFS-Ganesha server could make the operation fail in the checkpoint phase after the transaction had already been published. TransactionManager::commit() removes the transaction from activeTransactions before running the forced or automatic checkpoint. The context still held a non-owning activeTransaction pointer when the checkpoint threw. Error handling then attempted to roll back an object that the manager had already destroyed; connection teardown could repeat that rollback and let an exception escape a destructor, producing std::terminate/SIGABRT.

The same NFS outage can also fail page reads, dirty-page flushes, WAL cleanup, or shadow-file replay. Leaving a page locked, retaining a reserved frame, or deleting a WAL without syncing its directory can make the next open unrecoverable even after the original outage has ended.

How to reproduce?

1. build loopback file disk

sudo fallocate -l 4G /var/tmp/lbug-fsfreeze.img

sudo mkfs.ext4 -F /var/tmp/lbug-fsfreeze.img

sudo mkdir -p /mnt/lbug-fsfreeze

sudo mount -o loop /var/tmp/lbug-fsfreeze.img /mnt/lbug-fsfreeze

findmnt /mnt/lbug-fsfreeze\

2. mount NFS CLIENT

sudo apt update
sudo apt install -y nfs-ganesha nfs-ganesha-vfs
sudo systemctl disable --now nfs-server.service

sudo install -d -m 0755 /etc/ganesha

sudo cp -a /etc/ganesha/ganesha.conf /etc/ganesha/ganesha.conf.bak

sudo tee /etc/ganesha/ganesha.conf >/dev/null <<'EOF'
NFS_Core_Param {
Protocols = 4;
}

EXPORT {
Export_Id = 77;
Path = /mnt/lbug-fsfreeze;
Pseudo = /lbug-fsfreeze;
Access_Type = RW;
Squash = No_Root_Squash;
SecType = sys;
Protocols = 4;
Transports = TCP;

FSAL {
    Name = VFS;
}

}
EOF

sudo systemctl enable --now nfs-ganesha
sudo systemctl status nfs-ganesha --no-pager

sudo systemctl restart nfs-ganesha
sudo systemctl status nfs-ganesha --no-pager -l

sudo mkdir -p /mnt/lbug-nfs-client
sudo mount -t nfs4
-o vers=4.1,proto=tcp,hard,timeo=600,retrans=5
127.0.0.1:/lbug-fsfreeze /mnt/lbug-nfs-client

findmnt -T /mnt/lbug-fsfreeze -o TARGET,SOURCE,FSTYPE,OPTIONS
findmnt -T /mnt/lbug-nfs-client -o TARGET,SOURCE,FSTYPE,OPTIONS

3. use this test script to reproduce

python3 reconciler/cases/tc_local_lbug_copy_fsfreeze.py
--scenario nfs-node-copy-inflight-reopen
--mountpoint /mnt/lbug-nfs-client
--db-path /mnt/lbug-nfs-client/nfs-node-scenario-reopen.lbug
--freeze-seconds 180
--inflight-copy-nodes 1000000
--keep-db

or

python3 reconciler/cases/tc_local_lbug_copy_fsfreeze.py
--scenario nfs-copy-inflight
--mountpoint /mnt/lbug-nfs-client
--db-path /mnt/lbug-nfs-client/nfs-rel-scenario.lbug
--freeze-seconds 180
--inflight-copy-edges 1000000
--keep-db

tc_local_lbug_copy_fsfreeze.py

Are there known steps to reproduce?

No response