Production postmortemThe data corruption in the node.js stack

time to read 3 min | 415 words

A customer called us, complaining that RavenDB isn’t supporting internationalization. That was a big term to unpack. It boiled down to a simple issue. They were using Hebrew text in their system, consuming us from a node.js client, and they observed that sometimes, RavenDB would corrupt the data.

They would get JSON similar to this:

{ “Status”: "�", “Logged: true }

That… is not good. And also quite strange. I’m a native Hebrew speaker, so I threw a lot of such texts into RavenDB in the past. In fact, one of our employees built a library project for biblical texts, naturally all in Hebrew. Another employee maintained a set of Lucene analyzers for Hebrew. I think that I can safely say that RavenDB and Hebrew has been done. But the problem persisted. What was worse, it was not consistent. Every time that we tried to see what is going on, it worked.

We added code inside of RavenDB to try to detect what is going on, and there was nothing there. Eventually we tried to look into the Node.js RavenDB client, because we exhausted everything else. It looked okay, and in our tests, it… worked.

So we sat down and thought about what it could be. Let’s consider the actual scenario we have on hand:

  • Hebrew characters in JSON are being corrupted.
  • RavenDB uses UTF-8 encoding exclusively.
  • That means that Hebrew characters are using multi byte characters

That line of thinking led me to consider that the problem is related to chunking. We read from the network in chunks, and if the chunk happened to fall on a character boundary, we mess it up, maybe?

Once I started looking into this, the fix was obvious:


Here we go: ‍!

This bug is a great example of how things can not show up in practice for a really long time. In order to hit this you need chunking to happen in just the wrong place, and if you are running locally (as we usually do when troubleshooting), the likelihood you’ll see this is far lower. Given that most JSON property names and values are in the ASCII set, you need a chunk of just the right size to see it. Once we know about it, reproducing it is easy, just create a single string that is full of multi byte chars (such as an emoji) and make it long enough that it must be chunked.

The fix was already merged and released.

More posts in "Production postmortem" series:

  1. (05 Aug 2022) The allocating query
  2. (22 Jul 2022) Efficiency all the way to Out of Memory error
  3. (18 Jul 2022) Broken networks and compressed streams
  4. (13 Jul 2022) Your math is wrong, recursion doesn’t work this way
  5. (12 Jul 2022) The data corruption in the node.js stack
  6. (11 Jul 2022) Out of memory on a clear sky
  7. (29 Apr 2022) Deduplicating replication speed
  8. (25 Apr 2022) The network latency and the I/O spikes
  9. (22 Apr 2022) The encrypted database that was too big to replicate
  10. (20 Apr 2022) Misleading security and other production snafus
  11. (03 Jan 2022) An error on the first act will lead to data corruption on the second act…
  12. (13 Dec 2021) The memory leak that only happened on Linux
  13. (17 Sep 2021) The Guinness record for page faults & high CPU
  14. (07 Jan 2021) The file system limitation
  15. (23 Mar 2020) high CPU when there is little work to be done
  16. (21 Feb 2020) The self signed certificate that couldn’t
  17. (31 Jan 2020) The slow slowdown of large systems
  18. (07 Jun 2019) Printer out of paper and the RavenDB hang
  19. (18 Feb 2019) This data corruption bug requires 3 simultaneous race conditions
  20. (25 Dec 2018) Handled errors and the curse of recursive error handling
  21. (23 Nov 2018) The ARM is killing me
  22. (22 Feb 2018) The unavailable Linux server
  23. (06 Dec 2017) data corruption, a view from INSIDE the sausage
  24. (01 Dec 2017) The random high CPU
  25. (07 Aug 2017) 30% boost with a single line change
  26. (04 Aug 2017) The case of 99.99% percentile
  27. (02 Aug 2017) The lightly loaded trashing server
  28. (23 Aug 2016) The insidious cost of managed memory
  29. (05 Feb 2016) A null reference in our abstraction
  30. (27 Jan 2016) The Razor Suicide
  31. (13 Nov 2015) The case of the “it is slow on that machine (only)”
  32. (21 Oct 2015) The case of the slow index rebuild
  33. (22 Sep 2015) The case of the Unicode Poo
  34. (03 Sep 2015) The industry at large
  35. (01 Sep 2015) The case of the lying configuration file
  36. (31 Aug 2015) The case of the memory eater and high load
  37. (14 Aug 2015) The case of the man in the middle
  38. (05 Aug 2015) Reading the errors
  39. (29 Jul 2015) The evil licensing code
  40. (23 Jul 2015) The case of the native memory leak
  41. (16 Jul 2015) The case of the intransigent new database
  42. (13 Jul 2015) The case of the hung over server
  43. (09 Jul 2015) The case of the infected cluster