Random crashes of multichain daemon

+7 votes

We have Multichain network consisting of over 30 nodes. We are using multichain in version 2.3.3.
Unfortunately sometimes we observe unexpected crashes of daemon. Below logs from syslog:

Sep  1 13:56:54 Blockchain kernel: [280185.244489] bitcoin-dmsghan[1299]: segfault at 7f4554cdc82c ip 00000000004b823a sp 00007f45b4e1d950 error 6 in multichaind[400000+16e
8000]
Sep  1 13:56:54 Blockchain kernel: [280185.244555] Code: 01 00 00 8b 44 24 1c 89 84 24 50 01 00 00 48 8b 05 a3 c2 3c 02 48 8b b8 98 01 00 00 e8 3f e1 3a 00 4d 85 ff 0f 84 a
f 00 00 00 <41> 89 5f 24 e9 86 fd ff ff 48 8d 9c 24 60 01 00 00 48 8d 35 39 f4

And some last lines from multichain debug.log:

024-09-01 11:56:14 progress=1.000000  cache=0
2024-09-01 11:56:50 Accepted header for block 0000ce1ea85c7a7ca867a2891520dcd1954046742159e344c8522764e6e7325b found on alternative chain at height 5159405
2024-09-01 11:56:50 UpdateTip:            new best=0000cd0dd8140166c68bb856b4361916a1fc60a74afc18b4f1411792fb70f2da  height=5159406  log2_work=42.406208  tx=5890340  date=2
024-09-01 11:56:11 progress=1.000000  cache=0
2024-09-01 11:56:52 UpdateTip:            new best=00007ae05079b1cfd944dc04d5555d8a52993c4b6ddc2e785730337e8b4ea3b7  height=5159407  log2_work=42.406208  tx=5890341  date=2
024-09-01 11:56:16 progress=1.000000  cache=0
2024-09-01 11:56:52 Accepted header for block 0000cd83dc0a7d7c16389430e4987b1c3d8c828e479ffdb8ae67f6831434c541 found on alternative chain at height 5159407
2024-09-01 11:56:53 UpdateTip:            new best=0000a31b7fed3a818ddb72ddff38fb1a92a573fe4f1842f5b8e549ebbd9e9e73  height=5159408  log2_work=42.406208  tx=5890342  date=2
024-09-01 11:56:12 progress=1.000000  cache=0
2024-09-01 11:56:53 Accepted header for block 0000d0c2cf49dbf8447df139496a1758953c5fba2fd518e6e5a19496b81bb3c1 found on alternative chain at height 5159408
Using RdSeed as additional entropy source
Using RdRand as an additional entropy source

Do you have any idea what can be a reason for it? 

 

asked Sep 5 by manix

1 Answer

0 votes
The most likely problem is that the node's server is running out of memory because of how long the chain is (there's a small per-block memory footprint). We have a version 2.3.5 in testing which resolves this problem by removing that memory footprint. Please contact us if you'd like to try it.
answered Sep 5 by MultiChain
...