Multichain node hangs

Question

Some time after start a multichain node the RPC hangs. There is any way to know what it is doing?

I do multichain-cli, and I type getinfo and I will never respond:

MultiChain Core RPC client build 1.0 alpha 21 protocol 10005
Interactive mode

mincoins: getinfo
{"method":"getinfo","params":[],"id":1,"chain_name":"mincoins"}

I cannot stop the node. So I must do a Kill -9. Then if I restart again the node, I begins to respond again.

debug.log shows continuosly:

2016-05-28 11:41:00 ResendWalletTransactions()
2016-05-28 11:41:02 UpdateTip: new best=0000de164fda073a69fbe2f17b2c679da25e4e26f96ae865b2daea75f07c17ab  height=5755  log2_work=28.490873  tx=5764  date=2016-05-28 11:41:01 progress=1.000000  cache=125
2016-05-28 11:41:15 UpdateTip: new best=00008b69ce931b18d4850c75042fa06007c9cf34d21555c654c5ecf7b43ec14f  height=5756  log2_work=28.491124  tx=5765  date=2016-05-28 11:41:14 progress=1.000000  cache=126
2016-05-28 11:41:24 UpdateTip: new best=00006046c80864f7daec53f9b9bbcb5debcdc48891763dea95ed1f0a2b5666da  height=5757  log2_work=28.491374  tx=5766  date=2016-05-28 11:41:23 progress=1.000000  cache=127
2016-05-28 11:41:25 ResendWalletTransactions()

Thanks

Comments

Having the same issue (alpha 23). Any command I run hangs. This is the second chain I've created that hangs after some (probably less then 10 000) wallet transactions. Is there anything I can do to fix this (blockchain params or anything else)?

---

Please try this on alpha 24. If the problem persists, please stop the node, zip up the blockchain directory, then send it to us at multichain.debug at gmail, along with instructions on how we can reproduce the problem. We'll then take a detailed look and see if we can resolve the problem. Thanks!

---

I have a similar issue (alpha 28) after running encryptwallet. But for my case, I can't even start the node. Is there anything I can do to fix this node?

Thanks.

---

Hi John_cl, Can you please check this in beta 1, and if the problem persists, send the blockchain directory to us at multichain.debug at gmail to take a look?

---

It seems OK now. I ran out of space as wallet/txs.log ate 27GB. After increasing the space, I downloaded beta 1 and run it. I did not notice how long did it take, but it seems it added 3gb to the log. Now it runs normally (even after restart).

Thanks.

Answer 1

Thanks for reporting this. Can you please wait until it next hangs, kill it, and then .zip up the entire blockchain directory and send it to us. I'll sent you a private message with the email address.

Comments

Also it would help if you can run multichaind with the following additional options on the command line: -debug=mchn -debug=mchnminor -debug=net

---

Thanks, I killed the hanged nodes, and I have sent the tar file for complete directory. If I try now to start the nodes (I have three), I get the following:

MultiChain Core Daemon build 1.0 alpha 21 protocol 10005

MultiChain server starting
Chain mincoins already exists, adding 127.0.0.1:5001 to list of peers

New users can connect to this node using
multichaind mincoins@20.0.0.4:5003

: Corrupted block database detected.

Please restart multichaind with reindex=1.

I will recreate the chain (I have no problem because it is a test chain), and start the nodes with the parameters you have suggested.

Thanks,
   Fernando

---

Another thing, that I don't if it is related with this issue, is that I usally get errors:

'Transaction too large'

And I don't understand why, I increased máximum block size without success.

Thanks,
  Fernando

---

OK, thanks, we will look at the issues together.

---

I think it is related with the number of RPC connections we open to node.

I'm going to try the rpcthreads parameter to increase it.