Project

General

Profile

Port a Pipeline » History » Version 37

Joshua Randall, 09/05/2015 11:41 PM
adds '-r' option to `cp` when copying directory (fixes https://dev.arvados.org/issues/7186)

1 1 Nancy Ouyang
h1. Port a Pipeline
2
3
Like any other tool, Arvados requires time to learn. Thus, we don't encourage using Arvados for initial development of analysis pipelines or exploratory research on small subsets of data, when each quick-and-dirty iteration takes minutes on a single machine. But for any computationally-intense work, Arvados offers a lot of benefits.
4
5 8 Nancy Ouyang
Okay, cool, provenance, reproducibility, easily scaling to gigabytes of data and mucho RAM, evaluating existing pipelines like lobSTR quickly.
6 1 Nancy Ouyang
7
But what about if you want to these benefits when running your own pipelines?
8
In other words, how do you **port a pipeline** to Arvados?
9 35 Anonymous
10
{{toc}}
11 1 Nancy Ouyang
12
h2. 1. Quick Way
13
14
First, do you just want to parallelize a single bash script?
15
16 23 Nancy Ouyang
Check if you can use @arv-run@. Take this @arv-run@ example, which searches multiple FASTQ files in parallel, and saves the results to Keep through shell redirection:
17 9 Nancy Ouyang
18
    $ arv-run grep -H -n GCTACCAAGTTT \< *.fa \> output.txt
19 1 Nancy Ouyang
20 24 Nancy Ouyang
Or this example, which runs a shell script:
21
22
    $ echo 'echo hello world' > hello.sh    
23
$ arv-run /bin/sh hello.sh
24
25
(Lost? Check out http://doc.arvados.org/user/topics/arv-run.html)
26
27 1 Nancy Ouyang
h3. 1.1 Install arv-run
28
29 32 Brett Smith
(You can skip this step if you're working on an Arvados shell node.  @arv run@ is already installed and configured for you there.)
30
31 1 Nancy Ouyang
See: http://doc.arvados.org/sdk/python/sdk-python.html and http://doc.arvados.org/user/reference/api-tokens.html, or in short below:
32
<pre>
33 32 Brett Smith
$ sudo apt-get install python-pip python-dev python-yaml
34 1 Nancy Ouyang
$ sudo pip install --pre arvados-python-client
35
</pre>
36
(Lost? See http://doc.arvados.org/sdk/python/sdk-python.html )
37
38
If you try to use arv run right now, it will complain about some settings your missing. To fix that,
39
40 13 Nancy Ouyang
# Go to http://cloud.curoverse.com
41 1 Nancy Ouyang
# Login with any Google account (you may need to click login a few times if you hit multiple redirects from Google)
42
# Click in the upper right on your account name -> Manage Account
43 15 Nancy Ouyang
... !{height:10em}manage_account.png!:manage_account.png
44 1 Nancy Ouyang
# Optional: While you're here, click "send request for shell access", since that will give you shell access to a VM with all of the Arvados tools pre-installed.
45 14 Nancy Ouyang
1) !{height:10em}send_request.png!:send_request.png 2) !{height:10em}request_sent.png!:request_sent.png 3) !{height:10em}access_granted.png!:access_granted.png 
46 12 Nancy Ouyang
# Copy the lines of text into your terminal, something like
47 1 Nancy Ouyang
<pre>
48
HISTIGNORE=$HISTIGNORE:'export ARVADOS_API_TOKEN=*'
49
export ARVADOS_API_TOKEN=sekritlongthing
50 12 Nancy Ouyang
export ARVADOS_API_HOST=qr1hi.arvadosapi.com
51 1 Nancy Ouyang
unset ARVADOS_API_HOST_INSECURE
52 17 Nancy Ouyang
</pre> ... !{height:5em}terminal_ssh.png!:terminal_ssh.png
53 22 Nancy Ouyang
# If you want this to persist across reboot, add the above lines to @~/.bashrc@ or your @~/.bash_profile@
54 1 Nancy Ouyang
55
(Lost? See http://doc.arvados.org/user/reference/api-tokens.html )
56
57
h3. 1.2 Submit job to Arvados
58
59
First, check: Does your command work locally?
60
61 25 Nancy Ouyang
    $ grep -H -n TGGAAGT *.fa
62 1 Nancy Ouyang
63 25 Nancy Ouyang
... !{width:20em}grep-fasta.png!:grep-fasta.png
64
65
(If you want to follow along and don't have fasta files -- use the ones here: https://workbench.qr1hi.arvadosapi.com/collections/qr1hi-4zz18-0o2bt8216d7trrw)
66
67 1 Nancy Ouyang
If so, then submit it to arvados using @arv run@
68
69 25 Nancy Ouyang
    $ arv-run grep -H -n TGGAAGT \< *.fa \> output.txt
70 1 Nancy Ouyang
71 21 Nancy Ouyang
* This bash command stores the results as @output.txt@
72 25 Nancy Ouyang
* Note that due to the particulars of grep, Arvados will report this pipeline as **failed** if grep does not find anything, and no output will appear on Arvados
73 1 Nancy Ouyang
74
Your dataset is uploaded to Arvados if it wasn't on there already (which may take a while if it's a large dataset), your @grep@ job is submitted to run on the Arvados cluster, and you get the results in a few minutes (stored inside @output.txt@ in Arvados). If you go to Workbench at http://cloud.curoverse.com, you will see the pipeline running. It may take a few minutes for Arvados to spool up a node, provision it, and run your job. The robots are working hard for you, grab a cup of coffee.
75
76 21 Nancy Ouyang
(Lost? See http://doc.arvados.org/user/topics/arv-run.html )
77 25 Nancy Ouyang
78 1 Nancy Ouyang
h3. 1.3 However
79
80
If your pipeline looks more like
81
82 31 Nancy Ouyang
... !{width: 50%}https://arvados.org/attachments/download/428/provenance_graph_full.png!:https://arvados.org/attachments/download/428/provenance_graph_full.png
83
... _yes, that is a screenshot of an actual pipeline graph auto-generated by Arvados_
84 1 Nancy Ouyang
85 10 Nancy Ouyang
@arv-run@ is not powerful enough. Here we gooooo.
86 1 Nancy Ouyang
87 31 Nancy Ouyang
88 20 Nancy Ouyang
89 1 Nancy Ouyang
h2. 2. In Short
90
91
**Estimated reading time: 1 hour.**
92
93
You must write a **pipeline template** that describes your pipeline to Arvados.
94
95
h3. 2.1 VM (Virtual Machine) Access
96
97
Note: You'll need the Arvados set of command-line tools to follow along. The easiest way to get started is to get access to a Virtual Machine (VM) with all the tools pre-installed.
98
99 19 Nancy Ouyang
# Go to http://cloud.curoverse.com
100 1 Nancy Ouyang
# Login with google account (you may need to click login a few times, our redirects are not working well)
101 18 Nancy Ouyang
# Click in the upper right on your account name -> Manage Account
102
# Hit the "Request shell access" button under Manage Account in workbench.
103 1 Nancy Ouyang
104
h3. 2.2 Pipeline Template Example
105
106
Here is what a simple pipeline template looks like, where the output of program A is provided as input to program B. We'll explain what it all means shortly, but first, don't worry -- you'll never be creating a pipeline template from scratch. You'll always copy and modify an existing boilerplate one (yes, a template for the pipeline template! :])
107
108
109
    **pipelinetemplate.json**
110
    {
111
    "name": "Tiny Bash Script",
112
      "components": {
113
        "Create Two Files": {
114
          "script": "run-command",
115
          "script_version": "master",
116
          "repository": "nancy",
117
          "script_parameters": {
118
            "command": [
119
              "$(job.srcdir)/crunch_scripts/createtwofiles.sh"
120
            ]
121
          ,
122
          "runtime_constraints": {
123
            "docker_image": "nancy/cgatools-wormtable"
124
        ,
125
        "Merge Files": {
126
          "script": "run-command",
127
          "script_version": "master",
128
          "repository": "nancy",
129
          "script_parameters": {
130
            "command": [
131
              "$(job.srcdir)/crunch_scripts/mergefiles.sh",
132
              "$(input)"
133
            ],
134
            "input": {
135
              "output_of": "Create Two Files"
136
          ,
137
          "runtime_constraints": {
138
            "docker_image": "nancy/cgatools-wormtable"
139
          
140
h2. 3. simple and sweet port-a-pipeline example
141
142
Okay, let's dig into what's going on.
143
144
h3. 3.1 the setup
145
146
We'll port an artificially simple pipeline which involves just two short bash scripts, described as "A" and "B" below:
147
148
**Script A. Create two files**
149
Input: nothing
150 26 Nancy Ouyang
Output: two files (@out1.txt@ and @out2.txt@)
151 1 Nancy Ouyang
152
**Script B. Merge two files into a single file**
153
Input: output of step A
154 26 Nancy Ouyang
Output: a single file (@output.txt@)
155 1 Nancy Ouyang
156 27 Nancy Ouyang
Or visually (ignore the long strings of gibberish in the rectangles for now):
157 1 Nancy Ouyang
158 27 Nancy Ouyang
... !{height:30em}choose_inputs-small.png!:choose_inputs-small.png
159
160 26 Nancy Ouyang
161 1 Nancy Ouyang
Here's what we've explained so far in the pipeline template:
162
163
164
    **pipelinetemplate.json**
165
    {
166
    **"name": "Tiny Bash Script",**
167
      "components": {
168
       **"Create Two Files": {**
169
          "script": "run-command",
170
          "script_version": "master",
171
          "repository": "arvados",
172
          "script_parameters": {
173
            "command": [
174
              "$(job.srcdir)/crunch_scripts/ *createtwofiles.sh* "
175
            ]
176
          ,
177
          "runtime_constraints": {
178
            "docker_image": "nancy/cgatools-wormtable"
179
        ,
180
        **"Merge Files": {**
181
          "script": "run-command",
182
          "script_version": "master",
183
          "repository": "arvados",
184
          "script_parameters": {
185
            "command": [
186
              "$(job.srcdir)/crunch_scripts/ *mergefiles.sh* ",
187
              "$(input)"
188
            ],
189
           **"input": {**
190
              **"output_of": "Create Two Files"**
191
          ,
192
          "runtime_constraints": {
193
            "docker_image": "nancy/cgatools-wormtable"
194
195
h3. **3.2 arv-what?**
196
197
Before we go further, let's take a quick step back. Arvados consists of two parts
198
199
**Part 1. Keep** - I have all your files in the cloud!
200
201
You can access your files through your browser, using **Workbench**, or using the Arvados command line (CLI) tools (link: http://doc.arvados.org/sdk/cli/index.html ).
202
203 28 Nancy Ouyang
Visually, in Workbench, the built-in Arvados web interface, this looks like
204
... !{height:15em}port-a-pipeline-workbench-collection.png!:port-a-pipeline-workbench-collection.png
205 1 Nancy Ouyang
206 28 Nancy Ouyang
Or via the command-line interface
207
... !{height:10em}CLI-keep.png!:CLI-keep.png
208
209
210
211 1 Nancy Ouyang
**Part 2. Crunch** - I run all your scripts in the cloud!
212
213
Crunch both dispatches jobs and provides version control for your pipelines.
214
215
You describe your workflow to Crunch using **pipeline templates**. Pipeline templates describe a pipeline ("workflow"), the type of inputs each step in the pipeline requires, and . You provide a high-level description of how data flows through the pipeline—for example, the outputs of programs A and B are provided as input to program C—and let Crunch take care of the details of starting the individual programs at the right time with the inputs you specified.
216
217 28 Nancy Ouyang
... !{width:20em}provenance_graph_detail.png!:provenance_graph_detail.png
218
... _Each task starts when all its inputs have been created_
219 1 Nancy Ouyang
220
Once you save a pipeline template in Arvados, you run it by creating a pipeline instance that lists the specific inputs you’d like to use. The pipeline’s final output(s) will be saved in a project you specify.
221
222
Concretely, a pipeline template describes
223
224
* **data inputs** - specified as Keep content addresses
225
* **job scripts** - stored in a Git version control repository and referenced by a commit hash
226
* **parameters** - which, along with the data inputs, can have default values or can be filled in later when the pipeline is actually run
227
* **the execution environment** - stored in Docker images and referenced by Docker image name
228
229
**What is Docker**? Docker allows Arvados to replicate the execution environment your tools need. You install whatever bioinformatics tools (bwa-mem, vcftools, etc.) you are using inside a Docker image, upload it to Arvados, and Arvados will recreate your environment for computers in the cloud.
230
231 34 Bryan Cosca
**Protip:** Install stable external tools in Docker. Put your own scripts in a Git repository. This is because each docker image is about 1-5 GB, so each new docker image takes a while to upload (30 minutes) if you are not using Arvados on a local cluster. In the future, we hope to use small diff files describing just the changes made to Docker image instead of the full Docker image. [Last updated 19 Feb 2015]
232 1 Nancy Ouyang
233
h3. 3.3 In More Detail
234
235
Alright, let's put that all together.
236
237
    **pipelinetemplate.json**
238
    {
239
    "name": "Tiny Bash Script",
240
      "components": {
241
        "Create Two Files": {
242
          "script": "run-command",
243
          "script_version": "master",
244
          "repository": "nancy",
245
          "script_parameters": {
246
            "command": [
247
              "$(job.srcdir)/crunch_scripts/createtwofiles.sh" **#[1]**
248
            ]
249
          ,
250
          "runtime_constraints": {
251
            "docker_image": "nancy/cgatools-wormtable"
252
        ,
253
        "Merge Files": {
254
          "script": "run-command",
255
          "script_version": "master",
256
          "repository": "nancy",
257
          "script_parameters": {
258
            "command": [
259
              "$(job.srcdir)/crunch_scripts/mergefiles.sh", **#[2]**
260
              "$(input)"
261
            ],
262
            "input": {
263
              "output_of": "Create Two Files" **#[3]**
264
          ,
265
          "runtime_constraints": {
266
            "docker_image": "nancy/cgatools-wormtable"    
267
    
268
**Explanation**
269
    
270
[1] **$(job.srcdir)** references the git repository "in the cloud". Even though **run-command** is in nancy/crunch_scripts/ and is "magically found" by Arvados, INSIDE run-command you can't reference other files in the same repo as run-command without this magic variable.
271
272
Any output files as a result of this run-command will be automagically stored to keep as an auto-named collection (which you can think of as a folder for now).
273
274
[2] Okay, so how does the next script know where to find the output of the previous job? run-command will keep track of the collections it's created, so we can feed that in as an argument to our next script. In this "command" section under "run-command", you can think of the commas as spaces. Thus, what this line is saying is "run mergefile.sh on the previous input", or
275
276
  $ mergefiles.sh [directory with output of previous command]
277
278
[3] Here we set the variable "input" to point to the directory with the output of the previous command "Create Two Files".
279
280
(Lost? Try the hands-on example in the next section, or read more detailed documentation on the Arvados website: 
281
282
* http://doc.arvados.org/user/tutorials/running-external-program.html
283
* http://doc.arvados.org/user/topics/run-command.html
284
* http://doc.arvados.org/api/schema/PipelineTemplate.html )
285
286
h3. 3.4 All hands on deck!
287
288
Okay, now that we know the rough shape of what's going on, let's get our hands dirty.
289
290 2 Nancy Ouyang
*From your local machine, login to Arvados virtual machine*
291 1 Nancy Ouyang
292 4 Nancy Ouyang
Single step:
293 1 Nancy Ouyang
294 4 Nancy Ouyang
  nrw@ *@nrw-local@* $ ssh nancy@lightning-dev4.shell.arvados
295
296 1 Nancy Ouyang
(Lost? See "SSH access to machine with Arvados commandline tools installed" http://doc.arvados.org/user/getting_started/ssh-access-unix.html )
297
298
**In VM, create pipeline template**
299
300
A few steps:
301
302 30 Nancy Ouyang
  nancy@ *@lightning-dev4.qr1hi@* :~$ arv create pipeline_template
303 1 Nancy Ouyang
Created object qr1hi-p5p6p-3p6uweo7omeq9e7
304
$ arv edit qr1hi-p5p6p-3p6uweo7omeq9e7 #Create the pipeline template as described above! [[Todo: concrete thing]]
305
306
(Lost? See "Writing a pipeline template" http://doc.arvados.org/user/tutorials/running-external-program.html )
307
308
*In VM, set up git repository with run_command and our scripts*
309
310 33 Brett Smith
A few steps:
311 1 Nancy Ouyang
312 2 Nancy Ouyang
  $ mkdir @~@/projects
313
$ cd @~@/projects
314 33 Brett Smith
~/projects $ git clone git@git.qr1hi.arvadosapi.com:nancy.git
315 1 Nancy Ouyang
316 30 Nancy Ouyang
(Lost? Find your own git URL by going to https://workbench.qr1hi.arvadosapi.com/manage_account )
317 1 Nancy Ouyang
318 2 Nancy Ouyang
    ⤷Copy run_command & its dependencies into this crunch_scripts
319 33 Brett Smith
  $ git clone https://github.com/curoverse/arvados.git
320 1 Nancy Ouyang
321 2 Nancy Ouyang
(Lost? Visit https://github.com/curoverse/arvados )
322
323
  @  @$ cd ./nancy
324
  *@~/projects/nancy@* $ mkdir crunch_scripts
325
  *@~/projects/nancy@* $ cd crunch_scripts
326
  *@~/projects/nancy/crunch_scripts@* $ cp @~@/projects/arvados/crunch_scripts/run_command . #trailing dot!
327 37 Joshua Randall
  ~/projects/nancy/crunch_scripts$ cp -r ~/projects/arvados/crunch_scripts/crunchutil . #trailing dot!
328 2 Nancy Ouyang
329
  @  @$ cd ~/projects/nancy/crunch_scripts
330
331
  @  @$ vi createtwofiles.sh
332
    ⤷ $cat createtwofiles.sh
333
    #!/bin/bash
334
    echo "Hello " > out1.txt
335
    echo "Arvados!" > out2.txt
336
337
  @  @$ vi mergefiles.sh
338 1 Nancy Ouyang
    ⤷$cat mergefiles.sh
339 5 Nancy Ouyang
      #!/bin/bash *#[1]*
340
      PREVOUTDIR=$1 *#[2]*
341
      echo $TASK_KEEPMOUNT/by_id/$PREVOUTDIR *#[3]*
342 1 Nancy Ouyang
      cat $TASK_KEEPMOUNT/by_id/$PREVOUTDIR/*.txt > output.txt
343 33 Brett Smith
344 5 Nancy Ouyang
⤷ *Explanations*
345 6 Nancy Ouyang
*[1]* We use the @#!@ syntax to let bash know what to execute this file with
346 5 Nancy Ouyang
347
  ⤷To find the location of any particular tool, try using **which**
348
    $ which python
349
    /usr/bin/python
350
    $ which bash
351
    /bin/bash
352 33 Brett Smith
353 29 Nancy Ouyang
*[2]* Here we give a human-readable name, @PREVOUTDIR@, to the first argument (referenced using the dollar-sign syntax ala @$1@), given to @mergefiles.sh@, which (referring back to the pipeline template) we defined as the directory containing the output of the previous job (which ran @createtwofiles.sh@).
354
355
(Lost about @$1@? Google "passing arguments to the bash script").
356 33 Brett Smith
357 5 Nancy Ouyang
*[3]* Using the environmental variable @TASK_KEEPMOUNT@ allows us to not make assumptions about where **keep** is mounted. @TASK_KEEPMOUNT@ will be replaced by Arvados automatically with the name of the location to which **keep** is mounted on each worker node. (Lost? Visit http://doc.arvados.org/user/tutorials/tutorial-keep-mount.html )
358 33 Brett Smith
359 6 Nancy Ouyang
<pre>$ chmod +x createtwofiles.sh mergefiles.sh # make these files executable</pre>
360 1 Nancy Ouyang
361
**Commit changes and push to remote**
362
363 33 Brett Smith
A few steps:
364 1 Nancy Ouyang
365
  $ git status #check that everything looks ok
366 2 Nancy Ouyang
$ git add *
367 1 Nancy Ouyang
$ git commit -m “hello world-of-arvados scripts!”
368
$ git push
369 2 Nancy Ouyang
370 33 Brett Smith
**Create Docker image with Arvados command-line tools and other tools we want**
371 1 Nancy Ouyang
372 33 Brett Smith
> *Note:* This section assumes that you have Docker installed and usable under your user accounts.  However, because users with Docker access can defeat a lot of system security, it's not available on all Arvados shells.  If your Arvados VM doesn't provide you access to Docker, you have two options.  You can ask the site administrator to grant you access; or you can install Docker on your own GNU/Linux workstation, and upload the image to Arvados from there.  To learn how to do that, see the installation guides for "Docker Engine":https://docs.docker.com/ and the "Arvados Python SDK":http://doc.arvados.org/sdk/python/sdk-python.html, which includes the @arv-keepdocker@ tool to upload an image.
373 1 Nancy Ouyang
374
A few steps:
375
376 3 Nancy Ouyang
    $ docker pull arvados/jobs
377 36 Joshua Randall
$ docker run -ti -u root arvados/jobs /bin/bash
378 1 Nancy Ouyang
379 2 Nancy Ouyang
Now we are in the docker image.
380
381 33 Brett Smith
    root@4fa648c759f3:/# apt-get update
382 1 Nancy Ouyang
383 33 Brett Smith
    @  @⤷In the docker image, install external tools that you don't expect to need to update often.
384 1 Nancy Ouyang
    For instance, we can install the wormtable python tool in this docker image
385 4 Nancy Ouyang
    @  @# apt-get install libdb-dev
386 2 Nancy Ouyang
    @  @# pip install wormtable
387 1 Nancy Ouyang
388
    @  @  ⤷ Note: If you're installing from binaries, you should either
389 33 Brett Smith
        1) Install in existing places where bash looks for programs (e.g. install in /usr/local/bin/cgatools).
390 1 Nancy Ouyang
        To see where bash looks, inspect the PATH variable.
391 2 Nancy Ouyang
          #echo $PATH
392 3 Nancy Ouyang
          /usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin
393 1 Nancy Ouyang
        2) If you put them in a custom directory, remember them to reference them as such in your scripts
394
        (e.g. spell out /home/nrw/local/bin/cgatools).
395
        Arvados will not respect modifyng the $PATH variable by using the ~/.bashrc configuration file in the docker image.
396
397 6 Nancy Ouyang
(Lost? See http://doc.arvados.org/user/topics/arv-docker.html )
398 33 Brett Smith
399 6 Nancy Ouyang
  root@4fa648c759f3:/# exit
400 2 Nancy Ouyang
401 1 Nancy Ouyang
*Commit Docker image*
402
<pre>
403
$ docker commit 4fa648c759f3 nancy/cgatools-wormtable #Label the image thoughtfully
404 2 Nancy Ouyang
$ #For instance here I used the name of key tools I installed: cgatools & wormtable
405 1 Nancy Ouyang
</pre>
406 2 Nancy Ouyang
407 6 Nancy Ouyang
*Upload Docker image from your VM to Keep*
408 33 Brett Smith
409
> *Note:* @arv-keepdocker@ saves the Docker image in @~/.cache/arvados/docker@ before uploading, so it can resume in case of interruption.  If the @/home@ partition is not big enough to hold the Docker image, you may get strange I/O errors about pipe closed or stdin full.  You can prevent this by making @~/.cache/arvados/docker@ a symlink to another directory you control where enough space is available.  An example command for that might look like: @ln -s /scratch/MYNAME/docker ~/.cache/arvados/docker@
410
411 2 Nancy Ouyang
<pre>
412 33 Brett Smith
$ arv-keepdocker nancy/cgatools-wormtable #this takes a few minutes
413
$ arv-keepdocker #lists docker images in the cloud, so you can double-check what was uploaded </pre>
414 2 Nancy Ouyang
415 1 Nancy Ouyang
**Run this pipeline!**
416
Go to Workbench and hit **Run**.
417
<pre>$ firefox http://qr1hi.arvadosapi.com</pre>
418 30 Nancy Ouyang
[!image: workbench with 'tiny bash script']
419 1 Nancy Ouyang
420
*Note: If no worker nodes are already provisioned, your job may take up to 10 minutes to queue up and start.* Behind-the-scenes, Arvados is requesting compute nodes for you and installing your Docker image and otherwise setting up the environment on those nodes. Then Arvados will be ready to run your job. Be patient -- the wait time may seem frustrating for a trivial pipeline like this, but Arvados really excels at handling long and complicated pipelines with built-in data provenance and pipeline reproducibility.
421 33 Brett Smith
422 1 Nancy Ouyang
423
h3. 3.5 Celebrate
424
425
Whew! Congratulations on porting your first pipeline to Arvados! Check out http://doc.arvados.org/user/topics/crunch-tools-overview.html to learn more about the different ways to port pipelines to Arvados and how to take full advantage of Arvados's features, like restarting pipelines from where they failed instead of from the beginning. 
426
427
h2. 4. Debugging Tips and Pro-Tips
428
429
h3. **4.1 Pro-tips**
430
431
**Keep mounts are read-only right now. [19 March 2015]**
432
Need to 1) make some temporary directories or 2) change directories away from wherever you started out in but still upload the results to keep?
433
434
For 1, Explicitly use the $HOME directory and make the temporary files there
435
For 2, Use present working directory, $(pwd) at the beginning of your script to write down the directory where run-command will look for files to upload to keep.
436
437
Here's an example:
438
<pre>
439
$ cat mergefiles.sh
440
  TMPDIR=$HOME #directory to make temporary files in
441
  OUTDIR=$(pwd) #directory to put output files in
442
  mkdir $TMPDIR
443
  touch $TMPDIR/sometemporaryfile.txt #this file is deleted when the worker node is stopped
444
  touch $OUTDIR/someoutputfile.txt #this file will be uploaded to keep by run-command
445
</pre>
446
447
* make sure you point to the right repository, your own or arvados.
448
* make sure you pin the script versions of your python sdk, docker image, and script version or you will not get reproducibiltiy.
449
* if you have a file you want to use as a crunch script, make sure its in a crunch_scripts directory. otherwise, arvados will not find it. i.e. ~/path/to/git/repo/crunch_scripts/foo.py
450
451
h3. 4.2 Common log errors and reasons for pipelines to fail
452
453
Todo.
454
455
h3. 4.3 Miscellaneous Notes
456
457
Other ways to avoid the read-only keep mount problem is to use task.vwd which uses symlinks from the output directory which is writable to the colelction in keep. If you can change your working directory to the output directory and do all your work there, you'll avoid the keep read only issue.  (lost? see http://doc.arvados.org/user/topics/run-command.html )
458
    
459
When indexing, i.e. tabix, bwa index, etc. The index file tends to be created in the same directory as your fastq file. In order to avoid this, use ^. There is no way to send the index file to another directory. If you figure out a way, please tell me.
460
461
"bash" "-c" could be your friend, it works sometimes, sometimes it doesnt. I don't have a good handle on why this works sometimes.
462
463
if you're trying to iterate over >1 files using the task.foreach, its important to know that run-command uses a m x n method of making groups. I dont think I can explain it right now, but it may not be exactly what you want and you can trip over it. (lost? see http://doc.arvados.org/user/topics/run-command.html )
464
465
When trying to pair up reads, its hard to use run-command. You have to manipulate basename and hope your file names are foo.1 foo.2. base name will treat the group as foo (because you'll regex the groups as foo) and you can glob for foo.1 and foo.2. but if the file names are foo_1 and foo_2, you cant regex search them for foo becuase you'll get both names into a group and you'll be iterating through both of them twice, because of m x n. 
466
    
467
Your scripts need to point to the right place where the file is. Its currently hard to figure out how to grep the file names, you have to do some magic through the collection api.
468
469
h2. 5. Learn More
470
471
To learn more, head over to the Arvados User Guide documentation online: http://doc.arvados.org/user/