You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Geth version: 1.5.1
OS & Version: Linux
Commit hash : adc9fd1
Expected behaviour
The node with v1.4.16 upgraded to v1.5.1-alpha without any error.
Actual behaviour
The issue appears in testnet & mainnet env.
After the upgrade, the log reports unexpected journal version want 1 got 0, Head state missing, repairing, and "Rewound to block with state" number=46273779. It needs to wait for resync to chain tip.
Steps to reproduce the behaviour
Run a v1.4.16 node in testnet/mainnet, stop & upgrade binary to v1.5.1-alpha, restart the node.
it is as expected, v1.5.x introduce a new journal version:
// - Version 0: initial version
// - Version 1: storage.Incomplete field is removed
const journalVersion uint64 = 1
The new version is incompatible with previous version, v1.5.x can not support the journal format of v1.4.x as printed in log:
t=2024-12-10T03:13:26+0000 lvl=info msg="Failed to load journal, discard it" err="unexpected journal version want 1 got 0"
Then v1.5.x would discard the journal that was saved by v1.4.x, the journal mainly includes: 128 difflayers + [0, 256MB] trie nodes. Discard the journal would cause the node rewind to a previous flushed block height, the rewind block length depends on how many block's StateChangeSet were kept in this [0, 256MB] buffer, it is not a fixed length. We observed a testnet node rewinded 80k+ blocks and a mainnet node rewinded only 500+ block.
It would only occur only once after upgrading from v1.4.x to v1.5.x and take a while for the node to catch up.
System information
Geth version:
1.5.1
OS & Version: Linux
Commit hash : adc9fd1
Expected behaviour
The node with v1.4.16 upgraded to v1.5.1-alpha without any error.
Actual behaviour
The issue appears in testnet & mainnet env.
After the upgrade, the log reports
unexpected journal version want 1 got 0
,Head state missing, repairing
, and"Rewound to block with state" number=46273779
. It needs to wait for resync to chain tip.Steps to reproduce the behaviour
Run a v1.4.16 node in testnet/mainnet, stop & upgrade binary to v1.5.1-alpha, restart the node.
Backtrace
testnet:
mainnet:
When submitting logs: please submit them as text and not screenshots.
The text was updated successfully, but these errors were encountered: