Incorrect error message raised when underlying error is different Error: <Numer> appears to be an invalid tax name. This could be due to unexpected special characters.

[corneliusroemer]

Describe the bug
Using version 16.40.1, Linux x86_64 from conda-forge, I'm getting the error:

Error: 3048448 appears to be an invalid tax name.  This could be due to unexpected special characters.

Interestingly, I cant reproduce locally (macOS, ARM).

Command run: datasets download virus genome taxon 3048448 --no-progressbar --filename results/ncbi_dataset.zip --api-key None

Also getting this for other taxon ids, e.g. Error: 3052518 appears to be an invalid tax name. This could be due to unexpected special characters.

[corneliusroemer]

Here's with debug logs:

POST /datasets/v2/taxonomy/taxon_suggest HTTP/1.1
Host: api.ncbi.nlm.nih.gov
User-Agent: OpenAPI-Generator/1.0.0/go
Content-Length: 130
Accept: application/json
Content-Type: application/json
Ncbi-Phid: 18B066ABAA0DE0B4CC70A25C
X-Datasets-Client: datasets-cli
X-Datasets-Client-Arch: amd64
X-Datasets-Client-Cmd: download virus genome taxon 3052518 --no-progressbar --filename results/ncbi_dataset.zip --api-key None --debug
X-Datasets-Client-Os: linux
X-Datasets-Client-Version: 16.40.1
Accept-Encoding: gzip
{"exact_match":true,"tax_rank_filter":"higher_taxon","taxon_query":"3052518","taxon_resource_filter":"TAXON_RESOURCE_FILTER_ALL"}
Error: 3052518 appears to be an invalid tax name.  This could be due to unexpected special characters.
Use datasets download virus genome taxon <command> --help for detailed help about a command.

We're only seeing this on one server, is it possible you're actually doing 429 rate limiting but somehow reporting it as "invalid tax name"?

[theosanderson]

This does seem to be about an error message that doesn't always reflect the underlying situation

Repro:

datasets download virus genome taxon 3048448  --gateway-url https://nonexistent                     
Error: 3048448 appears to be an invalid tax name.  This could be due to unexpected special characters.

[RobertFalk]

Yes, that error currently is displayed when the taxon lookup fails to return anything, so it could occur for different problems as you demonstrated. I'll create a ticket to improve the error message. I was unfortunately not able to reproduce the error using 16.40.1 from conda-forge on my machine, and I don't see any results in our logging system based on the debugging information you provided. Maybe it failed to connect entirely? I can't say for sure. If it's still occurring, could you provide that same debug information, along with the time it was run?

[theosanderson]

(I'm on the same team as Cornelius) - we have now found we can't even connect w/ curl api.ncbi.nlm.nih.gov from the server in question, so it doesn't seem a datasets specific issue

[RobertFalk]

Thanks for the update - I created a ticket to improve that message so you'll better know what's going on if similar issues occur.

[corneliusroemer]

Thanks @RobertFalk! By the way, thanks for releasing the client code, it allowed us to look where this specific error message was raised:

datasets/client/apps/public/Datasets/v2/datasets/ResolveTaxons.go

Line 230 in 8a6e7b2

return nil, false, fmt.Errorf("%s appears to be an invalid tax name. This could be due to unexpected special characters.", taxName)

[aboffin]

Hi, I get a somewhat similar error:

When I tried to download using NCBI taxon id with or without --api-key option, I get either:

Error: Download error: stream error: stream ID 7; INTERNAL_ERROR; received from peer
Use datasets download genome taxon <command> --help for detailed help about a command.

or

Error: <taxonid> appears to be an invalid tax name.  This could be due to unexpected special characters.
Use datasets download genome taxon <command> --help for detailed help about a command.

Any pointers on how to resolve this is appreciated, thanks!

[ericcox1]

Hi @aboffin,

Thanks for your report. Could you please share the exact command that you used so that we can try to reproduce this?

Best,
Eric

[aboffin]

@ericcox1 the exact command was:

datasets download genome taxon "160490" --dehydrated --include protein --filename s_pyogenes_m1.zip --no-progressbar