A catalogue of bioinformatics failures traced to their real cause — by reading the pipeline's source, the config file, or the log the tool actually wrote. Nextflow, Snakemake, conda, Docker, samtools, GATK, R, HPC schedulers.
Every signature in the catalogue, so you can read the list without pasting anything. Each one opens the entry, with the cause and where the diagnosis came from.
command not found: mem You copied a line out of a pipeline log — that variable only exists inside the pipelineUnable to parse config file nextflow.config has a syntax error, usually an unbalanced brace or quoteNo such file or directory (in a task) The file exists on your machine but was never staged into the task's working directoryMissing output file(s) expected by process The task ran, but nothing matched the output pattern it declaredempty channel / no input received A glob matched nothing, so the process never ran at allterminated with an error exit status The workflow is relaying a tool failure — the reason is in the task's own error filemissing the QUALITY-column fasterq-dump is reading something that is not a real run accessionis a local non-kart file prefetch was handed a plain accession list, which it does not acceptchromosome naming mismatch (chr1 vs 1) Two files use different names for the same chromosomestruncated file / EOF marker absent The file was never finished being writtenfile not sorted / SO:coordinate The tool needs the file sorted in a specific order and this one is notfail to open file The path is wrong, the file is unreadable, or its index is missingsort: invalid option -- '@' That is the system sort, not samtools sort, because a variable in the command was empty$'\\r': command not found The script has Windows line endings, and Linux is reading the carriage return as part of the commandArgument list too long A glob expanded to more filenames than the shell can pass in one commandToo many open files The process hit the per-process limit on simultaneously open file handlesPermission denied (running a script) The file is not marked executable, or you are writing where you may notMissingOutputException The job reported success but Snakemake could not see the files it promisedDirectory cannot be locked A previous run died without releasing its lockAmbiguousRuleException Two rules can both produce the same file and Snakemake will not guessCyclicGraphException A rule ends up depending on its own outputTarget rules may not contain wildcards You asked for a rule by name, but that rule only makes sense per-sampleSolving environment: failed conda cannot find a combination of versions that satisfies everything at oncePackagesNotFoundError The package exists, but not on a channel you have — or not for your platformconda: command not found conda is installed but your shell has not been told about itCondaHTTPError / SSL error conda cannot reach the channel — network, proxy or certificate, not the packageexec format error The binary was built for a different CPU architecture than the machine running itNegativeArraySizeException An integer overflowed — this is not a memory problem, and more RAM will not fix it-Xmx set but ignored The heap setting is being silently ignored — check the file's line endingsOutOfMemoryError / Java heap space The JVM hit its heap ceiling — which is set on the JVM, not on the toolUnsupportedClassVersionError The tool needs a newer Java than the one running itCould not find artifact … in central The OSGi/Karaf container could not download its features, so its services never startedpermission denied /var/run/docker.sock Your user is not allowed to talk to the Docker daemonno space left on device The disk is full — often Docker's own storage rather than your home directorytool not found inside container The tool is in the image, but the container's PATH does not include it at runtimeoom-kill / exceeded memory limit The scheduler killed the job for exceeding the memory it requestedDUE TO TIME LIMIT The job hit its wall-clock limit and was cancelled mid-runexit status 137 / OOMKilled The process was killed from outside for using too much memorybwa mem prints its whole usage and exits 1 An option was written after the input files, and on macOS bwa counts it as extra input instead of an optionfail to locate the index files The aligner was given a reference that has no index beside itthere is no package called 'X' The package is installed somewhere, but not in the library this R session usescannot allocate vector of size R asked the OS for one contiguous block it could not getconnection refused / HTTP 522 Nothing is listening — this is a server or startup problem, not your installModuleNotFoundError after installing pip installed into a different Python than the one running the script