Project

General

Profile

GlideinwmsCondorAnnex » History » Version 8

Parag Mhashilkar, 05/04/2016 12:00 PM

1 1 Parag Mhashilkar
h1. GlideinwmsCondorAnnex
2 1 Parag Mhashilkar
3 3 Parag Mhashilkar
This is some documentation on very early version of condor_annex tools available from the HTCondor git repository.
4 3 Parag Mhashilkar
<pre>
5 3 Parag Mhashilkar
[parag@fermicloud338 condor_annex]$ git branch
6 3 Parag Mhashilkar
* V8_5-condor_annex-branch
7 3 Parag Mhashilkar
  master
8 3 Parag Mhashilkar
</pre>
9 1 Parag Mhashilkar
10 8 Parag Mhashilkar
h2. Evaluation
11 1 Parag Mhashilkar
12 8 Parag Mhashilkar
This is evaluation is based on very early version of condor_annex. HTCondor developers are already aware of some of the short comings and have plans to address them and improve its functionality at some point.
13 8 Parag Mhashilkar
14 8 Parag Mhashilkar
* *Need For Admin Privileges:* condor_annex does several privilaged operations. As a result the AWS user executing condor_annex needs to have admin like clearance. It is not clear that this is required and is a potential blocker for production style use. Different operations can be split into privileged operations and non privilieged operations. Privileged operations can e be run by the service/infrastructure provider to create the required environment, while non privileged operations that uses this environment can be used by a user to acquire resources.
15 8 Parag Mhashilkar
16 8 Parag Mhashilkar
* *Lack of ClassAds:* Glideinwms depends on information available from HTCondor to be in classad structure. This is not a strict requirement but info in classads makes it easier and be more as part of HTCondor ecosystem. It will be useful for condor_annex to periodically update the status of the annex and make it available as a classad in the Collector (?). This way all I deal with is HTCondor system and do not have to worry about the AWS components.
17 8 Parag Mhashilkar
18 8 Parag Mhashilkar
* *Usage Restricted to extend HTCondor pool only:* This use case though understandable makes it impossible to integrate within GlideinWMS. condor_annex expects HTCondor binaries to be baked into the VM. Annex through AWS functionalities will start up HTCondor Startd that report backs to a predefined collector. Annex will periodically monitor the state of the startds in the collector. In case of GlideinWMS, it brings its own (desired by VO) and starts up startd that reports to VO Collector. condor_annex is envisioned as a GlideinWMS factory side tool and may not have insight into the VO Collector. So making this functionality optional or easy to turn off is required from GlideinWMS point of view.
19 8 Parag Mhashilkar
20 8 Parag Mhashilkar
* *No Integration with condor_submit:* condor_annex is currently a standalone tool. Having the functionality to drive it through condor_submit -annex will be simplify things.
21 8 Parag Mhashilkar
22 8 Parag Mhashilkar
* condor_annex is developed using perl and requires aws client to be installed. Since aws client is available in python, maybe directly using aws-python APIs is also a viable option
23 8 Parag Mhashilkar
24 8 Parag Mhashilkar
25 8 Parag Mhashilkar
26 8 Parag Mhashilkar
27 8 Parag Mhashilkar
h2. Working Setup
28 8 Parag Mhashilkar
29 8 Parag Mhashilkar
h3. Preparations
30 8 Parag Mhashilkar
31 3 Parag Mhashilkar
* condor_annex requires --keypair. So had to run "aws configure". This created a $HOME/.aws directory with credentials in clear text. Its a ini file created by following command
32 1 Parag Mhashilkar
<pre>aws configure get region</pre>
33 3 Parag Mhashilkar
34 8 Parag Mhashilkar
h3. Working Command
35 3 Parag Mhashilkar
36 3 Parag Mhashilkar
<pre>
37 3 Parag Mhashilkar
/opt/condor/src/condor_annex/condor_annex \
38 3 Parag Mhashilkar
    --verbose \
39 3 Parag Mhashilkar
    --region=us-west-2 \
40 3 Parag Mhashilkar
    --project-id=annex_parag \
41 3 Parag Mhashilkar
    --instances=2 \
42 3 Parag Mhashilkar
    --expiry="2016-04-06 17:00" \
43 3 Parag Mhashilkar
    --central-manager=fermicloud385.fnal.gov \
44 3 Parag Mhashilkar
    --keypair=parag-annex \
45 3 Parag Mhashilkar
    --vpc=vpc-ed33af86 \
46 3 Parag Mhashilkar
    --subnet=subnet-ec33af87,subnet-e233af89,subnet-e333af88 \
47 3 Parag Mhashilkar
    --image-ids=ami-e826cd88 \
48 3 Parag Mhashilkar
    --spot-prices=0.06 \
49 3 Parag Mhashilkar
    --instance-types=m3.medium \
50 1 Parag Mhashilkar
    --password-file=/cloud/login/parag/wspace/glideinWMS/annex/password_file
51 1 Parag Mhashilkar
</pre>
52 3 Parag Mhashilkar
53 8 Parag Mhashilkar
h3. Internals of condor_annex
54 3 Parag Mhashilkar
55 1 Parag Mhashilkar
* Code is in perl and it invokes aws client commands.
56 3 Parag Mhashilkar
* Requires awscli python module
57 3 Parag Mhashilkar
<pre>pip install awscli</pre>
58 3 Parag Mhashilkar
59 8 Parag Mhashilkar
h3. Variables for Reference
60 3 Parag Mhashilkar
61 3 Parag Mhashilkar
$projectID (--project-id): Its an arbitrary string chosen by the user to handle future actions with this annex
62 3 Parag Mhashilkar
Make sure $annexSize is defined through --instances
63 3 Parag Mhashilkar
64 3 Parag Mhashilkar
$expiry (--expiry): When should this annex go away. Note the date format for the argument value.
65 3 Parag Mhashilkar
66 3 Parag Mhashilkar
$region (--region): If not provided use default region from ~/.aws/config. Use 'us-west-1' as hard default if all else fails.
67 3 Parag Mhashilkar
68 4 Parag Mhashilkar
$centralManager (--central-manager): Central manager where the condor startd will report to. Because this is required, we can not use condor_annex as is with GlideinWMS.
69 1 Parag Mhashilkar
70 4 Parag Mhashilkar
$passwordFile (--password-file): Password file created using condor_cred and used by condor startds/master in the VM to join the Condor Pool. Irrelevant in case of GlideinWMS.
71 3 Parag Mhashilkar
72 3 Parag Mhashilkar
$stackName (--stack-name): AWS stack name to use
73 3 Parag Mhashilkar
74 3 Parag Mhashilkar
$keypairName (--keypair): Name of the keypair in AWS to use while creating stack
75 3 Parag Mhashilkar
76 3 Parag Mhashilkar
$vpc (--vpc): VPC to use
77 3 Parag Mhashilkar
78 3 Parag Mhashilkar
$subnet (--subnet): Subnets to use
79 3 Parag Mhashilkar
80 3 Parag Mhashilkar
$imageIDList (--image-ids)
81 3 Parag Mhashilkar
82 1 Parag Mhashilkar
$spotPriceList (--spot-prices)
83 1 Parag Mhashilkar
84 3 Parag Mhashilkar
$instanceTypeList (--instance-types)
85 4 Parag Mhashilkar
86 5 Parag Mhashilkar
--
87 4 Parag Mhashilkar
88 1 Parag Mhashilkar
$s3Bucket="htcondor-annex-${safeCM}-${projectID}" where $safeCM is $centralManager after handling special characters like ':' and '.'
89 1 Parag Mhashilkar
90 4 Parag Mhashilkar
$passwordLocation: --password-location in s3 or $s3Bucket/brussel-sprouts
91 4 Parag Mhashilkar
92 4 Parag Mhashilkar
$configLocation: --config-location in s3 or $s3Bucket/basename($configFile)
93 4 Parag Mhashilkar
94 8 Parag Mhashilkar
h3. Workflow
95 4 Parag Mhashilkar
96 4 Parag Mhashilkar
+*VALIDATION PHASE*+
97 4 Parag Mhashilkar
98 3 Parag Mhashilkar
* Get the aws region to use.
99 3 Parag Mhashilkar
100 3 Parag Mhashilkar
* Create or get the stack to operate (modify/delete) on. If the stack does not exist it is created as needed using the keypair configured in AWS. If stack exists and --delete is given to the command, delete the stack. Use the VPC and Subnets passed by the user or use default with Name HTCondorAnnex. Since subnets are AZ specific, this is also a way to restrict annex to use a given AZ. Following AWS commands are used as part of various validations/information gathering in condor_annex. Either provide launch configuration ($launchConfigList) or provide the $imageIDList $spotPriceList $instanceTypeList
101 3 Parag Mhashilkar
102 3 Parag Mhashilkar
<pre>
103 3 Parag Mhashilkar
aws --region $region ec2 describe-key-pair
104 1 Parag Mhashilkar
aws --region $region ec2 describe-vpcs --filters 'Name=tag:Name,Values=HTCondorAnnex'
105 1 Parag Mhashilkar
aws --region $region ec2 describe-subnets --filters 'Name=tag:Name,Values=HTCondorAnnex' 'Name=vpc-id,Values=$vpc'
106 1 Parag Mhashilkar
</pre>
107 4 Parag Mhashilkar
108 4 Parag Mhashilkar
+*ACTION PHASE*+
109 4 Parag Mhashilkar
110 4 Parag Mhashilkar
* Create a s3 bucket to store $passwordFile and store it. If failed storing the password file to bucket, delete the bucket and roll back. Same action is performed for $configFile
111 4 Parag Mhashilkar
<pre>
112 4 Parag Mhashilkar
aws s3api create-bucket --acl private --bucket $s3Bucket
113 4 Parag Mhashilkar
aws s3 cp $passwordFile $passwordLocation
114 4 Parag Mhashilkar
aws s3 cp $configFile $configLocation
115 4 Parag Mhashilkar
</pre>
116 4 Parag Mhashilkar
117 1 Parag Mhashilkar
* Now create the cloud formation stack. $parameters below has info about the AIM ids, spot prices, instance types, VPCs, ProjectID, Subnet, ..., all the required stuff we got above.
118 5 Parag Mhashilkar
<pre>
119 5 Parag Mhashilkar
aws --region $region cloudformation create-stack \
120 5 Parag Mhashilkar
        --template-url "https://s3.amazonaws.com/condor-annex-${region}/template-${VERSION}" \
121 5 Parag Mhashilkar
        --stack-name $stackName --capabilities CAPABILITY_IAM --parameters $parameters
122 5 Parag Mhashilkar
</pre>
123 4 Parag Mhashilkar
124 4 Parag Mhashilkar
*CloudFormation launch configuration and Lambda requires high privileges at this time for following steps to work*
125 6 Parag Mhashilkar
!AWSPermissions-For-condor_annex.png!
126 4 Parag Mhashilkar
127 7 Parag Mhashilkar
* Create a autoscaling group if it does not exist. Wait for it to be created as we need its name to adjust the size and get the autoscaling group info by describe-stacks. For every stack in the output above DO SOMETHING if StackName matches our stack and StackStatus is CREATE_COMPLETE or UPDATE_COMPLETE.  Loop till the ResourceStatus is in CREATE_COMPLETE or UPDATE_COMPLETE for all the StackResources that are "AWS::AutoScaling::AutoScalingGroup"
128 4 Parag Mhashilkar
This is also a way for getting the autoscaling group names for future references
129 4 Parag Mhashilkar
130 4 Parag Mhashilkar
<pre>
131 4 Parag Mhashilkar
aws --region $region cloudformation describe-stacks
132 4 Parag Mhashilkar
aws --region $region cloudformation describe-stack-resources --stack-name $stackName
133 4 Parag Mhashilkar
</pre>
134 4 Parag Mhashilkar
135 1 Parag Mhashilkar
* Set the autoscaling group desired size that is computed to split the required annex-size across various autoscaling groups
136 1 Parag Mhashilkar
137 5 Parag Mhashilkar
<pre>
138 5 Parag Mhashilkar
aws --region $region autoscaling update-auto-scaling-group \
139 5 Parag Mhashilkar
       --auto-scaling-group-name $asgName --max-size $size --desired-capacity $size
140 5 Parag Mhashilkar
</pre>
141 4 Parag Mhashilkar
142 4 Parag Mhashilkar
* Once the annex has been created set/update the expiration time. This is some complicated code not worth describing at this time. In short it depends on heartbeat and alarms.
143 4 Parag Mhashilkar
144 4 Parag Mhashilkar
* Determine how big annex has grown and if we are at the required capacity
145 4 Parag Mhashilkar
146 4 Parag Mhashilkar
* Wait for the annex nodes to join the HTCondor pool. This is a BUMMER because we do not want to use annex in this mode and want a means to skip this. Also it is using condor_status -constraint 'ProjectID=="$projectID"' which means that it will conflict with the generic projectid classad attribute.