I encountered an error while running Snippy with a GenBank reference and two raw reads. Any suggestions? what I should do …?
Error is denote as “It used more memory than it was allocate” .
(Galaxy.org)
This message can be confusing! We can help here.
When a tool fails with that error message, it usually means that there is a data input or parameter problem to address. It is not related to the data storage space in your account, but instead the memory allocated to tools on the public clusters.
Something probably went wrong, the tool was confused, memory usage soared outside of expected bounds, and the tool died. 
This is pretty common behavior when using computational tools, especially with new data or if the tool itself is new to you.
How to troubleshoot?
-
Inspect your data. That can mean running QA on the reads and verifying that the custom genome is formatted in a way the target tool can understand.
-
Galaxy hosts tools created by a wide range of open source developers, and each can have a slightly different idea of what a “basic format” for a reference genome looks like.
- The guide has instructions that can help you to quickly simplify format. → FAQ: How to use Custom Reference Genomes?
-
For read QA, there are a few different methods. To find resources, review the tool panel, or search our training site.
- GTN search → GTN Materials Search (query=quality) → Hands-on: Quality Control / Quality Control / Sequence analysis
-
-
Learn more about using the tool. If there is a Galaxy tutorial, find that on tool forms down in the Help section. You’ll also find links to the original tool documentation (best for scientific tuning) then abbreviated help focused on how to translate general usage onto the Galaxy form.
-
Analysis job really is too large to process (rare!). That could be at one public server or all. This situation is less common but certainly possible!
-
You can try at a different public Galaxy server to see what happens. Each hosts distinct public computational resources (and reference data). → FAQ: Transfer entire histories from one Galaxy server to another
-
Galaxy has private server options for really large or batch work.
-
What to do
-
Review your job details. Where to look → FAQ: Troubleshooting errors
-
Compare your usage to examples, run some QA, adjust reference data, maybe adjust parameters, and finally consider trying at a different UseGalaxy server.
-
If you would like help with any of this, you can share your work for feedback here. See → How to get faster help with your question. You can post back screenshots of the job details or post back the history share link.
Let’s start there and let us know if you need more help or are able to solve this yourself! 