-
Notifications
You must be signed in to change notification settings - Fork 149
neo4j-driver causing Node.js process to crash silently #324
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Comments
Hi @acjohnso25, We have never seen problem like this before. Which version of Node are you using? |
We are using Node.js version 4.8.0. Our Neo4j cluster is running 3.2.7. The possibility of running out of memory occurred to me previously, so I already hacked in a bit of monitoring -- basically spitting out the results of I suppose it's possible that something goes haywire inside that 60s interval so it's not showing up there. However, we always run four instances of this process on each machine, so if one was suddenly chewing up all the resources, I'd expect to see all four (or at least more than one) crash around the same time...but this has not happened once. I'll post the complete crash report in a separate comment. |
BTW, the crash happened again this morning, on a different machine. The stack trace from that crash is basically the same as the one I pasted initially. |
Is that a full crash dump? It looks cut at the end. |
Hm...perhaps. That was what was on the machine; the only thing I removed were some additional logs at the end that contained our service and hostnames. I'll find and post another one... |
Here are the dependency versions installed on one of our production instances, from
|
@acjohnso25 thanks for uploading the file. Stacktrace points to transpiled code and ends in driver's var _possibleConstructorReturn2 = require('babel-runtime/helpers/possibleConstructorReturn'); it's only babel's code from then on. We have not seen such errors before, so they might be environment dependent. Could you please increase memory limit for Node to 4GB? I know you have monitoring and we see nothing about memory in the crashdump, but increasing heap will help us to completely rule out memory usage issue. Is upgrading Node an option for you? |
@lutovich Neither of those is trivial to do in our production environment (which is where we are seeing this) at the moment. I'll provide an update if we get the chance to do that. I'm taking a look through babel-runtime issues to see if there are any similar reports there...but there are 460 issues open...which is a lot... :( |
@lutovich Is there an older version of the driver that we could try, prior to it using babel-runtime? I've tried to figure out what release incorporated this change, but github's UI doesn't seem to make this easy. The best I've been able to figure out is that babel-runtime was added with this commit: e6c5ce1 But figuring out what version that commit wound up in is not obvious to a GH neophyte like me... (I've used Bitbucket more.) Update: I found the PR, #162, which suggests the change was merged into 1.1 branch. |
Hi @acjohnso25, Did you have a chance to try the pre-babel driver? |
Hi @acjohnso25, Do you still experience this issue? Did you try newer Node version or maybe have more info about how can we reproduce the problem? |
Hi @lutovich, sorry for not getting back to you last time. Busy couple of weeks. I never did try the pre-babel driver; it appeared that I'd have to step back to 1.0 or a 0.x release to do so, which I figured would probably introduce all kinds of other problems.
Sorry, I don't have any more info on reproducing the problem at this time. |
Close due to lack of information. Pls be free to reopen it once more information is available. |
We're using neo4j-driver 1.5.2 in a Node.js service and have been seeing mysterious crashes for months now. We tried catching uncaught exceptions and even wrapping
process.exit()
to try to track them down, but no luck; the process always just dies silently. Finally we decided to turn on some profiling and saw this on the next crash:I'm not really sure what's going on here, or even how to reproduce it. It's very intermittent, doesn't seem to happen in any predictable way. (Believe me, I've tried.)
If anybody needs more info or has ideas for ways to repro, I'm all ears.
The text was updated successfully, but these errors were encountered: