# Rel receipe is getting killed after a min or so. **URL:** \ **Category:** Uncategorized\ **Created:** [September 22, 2022, 8:07pm UTC](https://support.prodi.gy/t/rel-receipe-is-getting-killed-after-a-min-or-so/5961 "2022-09-22T20:07:28Z")\ **Posts on this page:** 4\ **Page:** 1
**Author:** ![monWork](https://sea2.discourse-cdn.com/flex020/user_avatar/support.prodi.gy/monwork/32/3214_2.png) [@monWork](https://support.prodi.gy/u/monWork)\ **Post date:** [September 22, 2022, 8:07pm UTC](https://support.prodi.gy/t/rel-receipe-is-getting-killed-after-a-min-or-so/5961/1 "2022-09-22T20:07:28Z")
![image](https://us1.discourse-cdn.com/flex020/uploads/prodigy/original/2X/4/47aaa70974cbc41d99fbe721f812b5f05a2e7aee.png) So i have this rel manual receipe to manually annotate spans and relations and my dataset size is just 10mb, so i dont know its getting killed with this error of IOSTREAM.flush time out. ---
**Author:** ![ryanwesslen](https://sea2.discourse-cdn.com/flex020/user_avatar/support.prodi.gy/ryanwesslen/32/2969_2.png) [@ryanwesslen](https://support.prodi.gy/u/ryanwesslen)\ **Post date:** [September 23, 2022, 8:54am UTC](https://support.prodi.gy/t/rel-receipe-is-getting-killed-after-a-min-or-so/5961/2 "2022-09-23T08:54:22Z")
hi @monWork! This seems like a memory issue with iPython (Jupyter): [Memory Error: IOStream.Flush timed out · Issue #334 · ipython/ipykernel · GitHub](https://github.com/ipython/ipykernel/issues/334) Are you able to run the same command but from a terminal (not a Jupyter notebook)? As you may have found too, I found [this StackOverflow post](https://stackoverflow.com/questions/70580988/iostream-flush-timed-out-what-does-it-mean) for a similar issue. The proposed solution was running: ```python %reset -f ``` ---
**Author:** ![monWork](https://sea2.discourse-cdn.com/flex020/user_avatar/support.prodi.gy/monwork/32/3214_2.png) [@monWork](https://support.prodi.gy/u/monWork)\ **Post date:** [September 27, 2022, 6:57pm UTC](https://support.prodi.gy/t/rel-receipe-is-getting-killed-after-a-min-or-so/5961/3 "2022-09-27T18:57:19Z")
Hi @ryanwesslen I tried working it through the terminal as well and still the same, the process gets killed after a while. However, one thing i wanted to notify you is that the annotation examples are whole documents with good length. Could that be the reason for it ? I do see this warning `⚠ Long example with 588 tokens detected. This can potentially lead to slower rendering and annotation in the web app. Consider splitting your texts into smaller chunks or sentences.` ---
**Author:** ![ryanwesslen](https://sea2.discourse-cdn.com/flex020/user_avatar/support.prodi.gy/ryanwesslen/32/2969_2.png) [@ryanwesslen](https://support.prodi.gy/u/ryanwesslen)\ **Post date:** [September 27, 2022, 7:20pm UTC](https://support.prodi.gy/t/rel-receipe-is-getting-killed-after-a-min-or-so/5961/4 "2022-09-27T19:20:57Z")
hi @monWork ! > [@monWork](#): > > However, one thing i wanted to notify you is that the annotation examples are whole documents with good length. Could that be the reason for it ? > > I do see this warning `⚠ Long example with 588 tokens detected. This can potentially lead to slower render` Very likely. This is a well known issue. > [@relation is responding very slowly](https://support.prodi.gy/t/relation-is-responding-very-slowly/4732/2): > > Hi and sorry about this – it's currently expected that the interface becomes less performant for very long documents with many tokens, and we're working on a rewrite that doesn't have this problem. I think what makes it additionally tricky in your case is that you just end up with more tokens overall, due to the way the characters are segmented. As a workaround, one thing you can do is use patterns to disable all tokens that you know won't ever be part of a relation – of course, only if that's … > [@UI slow when serving snippets over 500 words](https://support.prodi.gy/t/ui-slow-when-serving-snippets-over-500-words/4342/2): > > Hi! Which annotation interface are you using? Are you using the relations UI? If so, that can currently become slower with larger documents (the specifics also depend on your browser and machine). I definitely want to work on improving the performance here – the tricky part is that the interface is quite complex so we might need a better way to represent the tokens and spans (which technically all have to be able to be connected to any other token or span). If you're using any other text-based … Try [disabling tokens](https://prodi.gy/docs/dependencies-relations/#example-bio) (`--disable-patterns`) for terms you don't need for labeling (this will speed up some). Or alternatively, could you break up your documents into sentences? Let me know if you need help