Add new troubleshooting page on why a process is retrying

sites/docs/src/content/docs/usage/troubleshooting/overview.md

@@ -17,6 +17,7 @@ parentWeight: 20
 1. [Unable to acquire lock error](/docs/usage/troubleshooting/aquire_lock_error)
 1. [Docker permission errors](/docs/usage/troubleshooting/docker_permissions)
 1. [IPv6 network errors](/docs/usage/troubleshooting/ipv6)
+1. [Processes are retrying](/docs/usage/troubleshooting/retries.md)
 
 ## How to use these pages
 

sites/docs/src/content/docs/usage/troubleshooting/retries.md

@@ -0,0 +1,15 @@
+## Processes are retrying
+
+### Why did processes report an error but then retry?
+
+One of the nice things about Nextflow is it offers the ability to retry processes if they encounter an error and fail with certain bash [exit status](https://en.wikipedia.org/wiki/Exit_status) or codes.
+Some of these errors have common causes, which allows us to provide solutions to these problem.
+
+A common issue is a tool requiring more memory than is (initially) made available to it based on the default memory specifications set in the pipeline.
+Such errors (out of memory, or OOM) are often identified by exit code `104`, and those falling between `130` - `145`.
+
+Therefore all nf-core pipelines [by default will retry](https://github.com/nf-core/tools/blob/930ece572bf23b68c7a7c5259e918a878ba6499e/nf_core/pipeline-template/conf/base.config#L18) a process if it hits one of those exit codes, but requesting more resources (memory, CPUs, and time) for the re-submitted job.
+
+All other exit codes will cause the pipeline to fail immediately, and will not be retried.
+
+However some pipelines may extend this list, or provide different retry conditions, based on the behaviour of the specific tools in the used in the pipeline.