docker: Error parsing reference: ":8787" is not a valid repository

[dncnbrn]

Using the default example:

docklet_create() %>%
  docklet_rstudio()

I get this every time:

Using default ssh keys: id_rsa
NB: This costs $0.01488 / hour until you droplet_delete() it
Waiting for create ........................
Error: No network interface registered for this droplet
  Try refreshing like: droplet(d$id)

Not ideal, but if I try to refresh:

d <- droplets()
d[[1]]$id %>% docklet_rstudio()

It then runs through the docker process, but ends here:

docker: Error parsing reference: ":8787" is not a valid repository/tag: invalid reference format.
See 'docker run --help'.
Error: ssh failed
ssh -o BatchMode=yes -o StrictHostKeyChecking=no -o UserKnownHostsFile=/var/folders/nk/vwrxh1rj72b2tflmmqnd5kqw0000gq/T//Rtmp4hGAQj/hosts root@206.189.30.63 'docker run  -d  -p  8787 :8787    -e USER= rstudio  -e PASSWORD= rstudio  -e EMAIL= rstudio@example.com   rocker/rstudio  '

Re the ssh failed, I can ssh into the droplet from terminal. Also worth noting that the plumber do_provision() function seems to run without any problems.

Session Info

R version 3.5.1 (2018-07-02)
Platform: x86_64-apple-darwin15.6.0 (64-bit)
Running under: macOS  10.14

Matrix products: default
BLAS: /System/Library/Frameworks/Accelerate.framework/Versions/A/Frameworks/vecLib.framework/Versions/A/libBLAS.dylib
LAPACK: /Library/Frameworks/R.framework/Versions/3.5/Resources/lib/libRlapack.dylib

locale:
[1] en_GB.UTF-8/en_GB.UTF-8/en_GB.UTF-8/C/en_GB.UTF-8/en_GB.UTF-8

attached base packages:
[1] stats     graphics  grDevices utils     datasets  methods  
[7] base     

other attached packages:
[1] analogsea_0.6.0

loaded via a namespace (and not attached):
 [1] Rcpp_1.0.0          cellranger_1.1.0   
 [3] compiler_3.5.1      later_0.7.5        
 [5] base64enc_0.1-3     tools_3.5.1        
 [7] aws.s3_0.3.12       digest_0.6.18      
 [9] packrat_0.4.9-3     jsonlite_1.5       
[11] lattice_0.20-35     rstudioapi_0.8     
[13] curl_3.2            yaml_2.2.0         
[15] httr_1.3.1          xml2_1.2.0         
[17] knitr_1.20          stats4_3.5.1       
[19] lmtest_0.9-36       grid_3.5.1         
[21] nnet_7.3-12         data.table_1.11.6  
[23] R6_2.3.0            flexmix_2.3-14     
[25] readxl_1.1.0        Formula_1.2-3      
[27] plumber_0.4.6       magrittr_1.5       
[29] promises_1.0.1      modeltools_0.2-22  
[31] MASS_7.3-50         aws.signature_0.4.4
[33] httpuv_1.4.5        betareg_3.1-1      
[35] sandwich_2.5-0      mfx_1.1            
[37] stringi_1.2.4       crayon_1.3.4       
[39] zoo_1.8-4  

[sckott]

sorry about the problems there @dncnbrn

can you install the dev version and try again?

remotes::install_github("sckott/analogsea")

[dncnbrn]

Thanks @sckott - this has solved the problem; it no longer gives the error message, although when it then proceeds to open the droplet on port 8787 I just get 'The site can't be reached' and 'ERR_CONNECTION_REFUSED' on Chrome? (Apologies if this a settings issue I need to fix.)

[sckott]

sounds like there was a problem there. can you give more details of the code used and any error messages

[dncnbrn]

Pretty much as before, but no error messages given:

library(analogsea)
docklet_create(region="lon1") 
d <- droplets()
d[[1]]$id %>% docklet_rstudio()

On the last step, I get New server fingerprint: ..., then Using default tag: latest, then latest: Pulling from rocker/rstudio, then it runs through the Docker instructions, ending with Digest: sha256... and finally Status: Downloaded newer image for rocker/rstudio:latest..., before opening up a tab on Chrome with the IP address and then :8787 appended to it, and Chrome just gives back ERR_CONNECTION_REFUSED.

So it seems like everything is working fine but then it doesn't connect to the session. Added: I've been able to get access to a DO server on port 8787 if I create a droplet and then build an R and RStudio setup through the terminal.

[sckott]

thanks @dncnbrn for the details. I'll see if I can replicate this problem somehow and get back to you

[sckott]

related #170

[dncnbrn]

That works for me - changing from the default password has solved the problem.

[sckott]

@dncnbrn install remotes::install_github("sckott/analogsea") and try again, should work now.

user and password are now required parameters, and rstudio can't be 'rstudio'