sebdah / dynamic-dynamodb Goto Github PK

View Code? Open in Web Editor NEW

617.0 617.0 124.0 1.59 MB

Dynamic DynamoDB provides auto scaling for AWS DynamoDB

Home Page: http://dynamic-dynamodb.readthedocs.org/

License: Apache License 2.0

Makefile 0.06% Python 99.94%

dynamic-dynamodb's People

Contributors

Stargazers

Watchers

Forkers

pragnesh carrot-garden rsharmachamp willejs dopplesoldner tmorgan leeyingmu msh9 abinesh ddani neonbunny joshlk suzuken robbwagoner gabhi johnbartels tayl0r jofen ulsa tellybug tpurtell matiasjrossi rezanachmad localytics danilop bellpeterm sp1tf1r3 kps shaharck whiteboardmonk sincntx mavrick nickveenhof ricardclau bsideup acquiachrisnagy finoccf lcabral37 sysalexis priyankag1905 suzaku benweet jasonrdsouza omnidavesz foroozan venkatanathen tcana1 lead-id superprat qqshfox kylealandavis kcabral zaquaz lukewaite 4ossiblellc kennethjor mindaugasrukas cundohubs lhamregit erwinverdonk dhoeric luminalhq avadlapudi zdoherty naomichi-y xavierbruguera booyildirim amalnanavati erikljung omar-shibli-zocdoc moberg fingerfoodstudios optionalg remitly wrenth04 benhigginbottom robinmisfit killdash9 windgo nitin608 michaelkaye jerelim pilgrim2go kill3rbee rbramwell jessyli xilongfan spokeo tantra35 eis-datascience smartsensebydigi amalaannrose etsangsplk danielstirk bcuff byron70 cjy0125 dengyuan1989 opteemister utkarshd420

dynamic-dynamodb's Issues

Send e-mail when DynamoDB limits are reaches (e.g. 2 scale down's per day)

Send e-mail when DynamoDB limits are reaches (e.g. 2 scale down's per day).

Cloudwatch monitor adjustment

We are currently using dynamic-dynamo in our production servers, and due the behaviour of our systems, dynamic-dynamo is continuously changing the provisioning resources asigned to our DynamoTables.
However, we are seeing something really weird: the configured cloudwatch alarm limit also changes when the provisioned throughput is changed.
For example, we have a table provisioned originally with 100 reads, with an alarm configured with a 80% of the capacity. This table has been updated up and down for some days, and now the alarm is configured to the 27%.

We don't know yet if this is an issue from AWS or from dynamic dynamo, but in both cases it would be really nice if Dynamic Dynamo could set the alarm in the original status.

Support multiple tables in one configuration file

It would be great if Dynamic DynamoDB could handle more than one table at a time.

Change pid file name

The pid file name is badly named, update it to something that makes sense

Handle combinations of configuration file and command line options better

Today Dynamic DynamoDB is more or less handling configuration files OR command line options. This should be changed so that combinations are supported. I think that command line options should have precedence over configuration file options.

No logging in --daemon mode

When dyn-dyn is started in daemon mode, it does not update the logfile. It works when started interactively.

Add --version flag to dynamic-dynamodb command

Add --version flag to dynamic-dynamodb command

Split main function to separate Python file

Split main function to separate Python file so the DynamicDynamoDB class gets its own file.

Support for daemonizing Dynamic DynamoDB

Handle DynamoDB exceptions when updating throughput

Handle DynamoDB exceptions when updating throughput.

When scaling up more than >100%
When scaling down more than twice per day

Increasing to a minimum provisioned throughput value doesn't take into account the current table's throughput

Say for instance I have a table with a write throughput of 2, and in my configuration, that table's minimum write throughput is set to 5. When attempting to increase to that minimum write throughput, amazon will throw an error, because you can only increase table's throughput up to twice the amount of the current throughput. Might I suggest something along the lines of:

return min( get_table_option(table_name, 'min_provisioned_reads'), (current_provisioning * 2) )

return min( get_table_option(table_name, 'min_provisioned_writes'), (current_provisioning * 2) )

When returning a minimum provisioned read or write value so that the table can at least get closer to it's minimum provisioned value.

I suppose this is an edge case, but I'd rather give you the option to ignore it instead of the issue never being heard.

Mix up between percent and units

Mix up between percent and units when getting the consumed read units.

wrong claim that "Scaling limit exeeded"

I just started dyn-dyn on a set of table that was manually configured to 2000-5000 CU per table (for a migration job). The actual usage now will be more in the 5-15 CU range.

dynamic-dynamodb fails on many of these tables as below:

2013-05-21 07:33:03,880 - dynamic-dynamodb - INFO - PROD_AccountRelation - Consumed read units: 9%
2013-05-21 07:33:03,921 - dynamic-dynamodb - INFO - PROD_AccountRelation - Consumed write units: 1%
2013-05-21 07:33:03,921 - dynamic-dynamodb - INFO - PROD_AccountRelation - Changing provisioning to 550 read units and 550 write units
2013-05-21 07:33:04,185 - dynamic-dynamodb - INFO - Provisioning updated
2013-05-21 07:33:04,345 - dynamic-dynamodb - INFO - PROD_PostrefsRepostLookup - Consumed read units: 1%
2013-05-21 07:33:04,480 - dynamic-dynamodb - INFO - PROD_PostrefsRepostLookup - Consumed write units: 2%
2013-05-21 07:33:04,480 - dynamic-dynamodb - INFO - PROD_PostrefsRepostLookup - Changing provisioning to 85 read units and 42 write units
2013-05-21 07:33:04,853 - dynamic-dynamodb - INFO - Provisioning updated
2013-05-21 07:33:04,957 - dynamic-dynamodb - INFO - PROD_ReactionUserRepost - Consumed read units: 7%
2013-05-21 07:33:05,007 - dynamic-dynamodb - INFO - PROD_ReactionUserRepost - Consumed write units: 4%
2013-05-21 07:33:05,007 - dynamic-dynamodb - INFO - PROD_ReactionUserRepost - Changing provisioning to 22 read units and 22 write units
2013-05-21 07:33:05,252 - dynamic-dynamodb - INFO - Provisioning updated
2013-05-21 07:33:05,305 - dynamic-dynamodb - INFO - PROD_OverallCounts - Consumed read units: 2%
2013-05-21 07:33:05,342 - dynamic-dynamodb - INFO - PROD_OverallCounts - Consumed write units: 2%
2013-05-21 07:33:05,342 - dynamic-dynamodb - INFO - PROD_OverallCounts - Changing provisioning to 33 read units and 33 write units
2013-05-21 07:33:05,370 - dynamic-dynamodb - WARNING - PROD_OverallCounts - Scaling limit exeeded. The table can only be scaled down twice per day.
2013-05-21 07:33:05,430 - dynamic-dynamodb - INFO - PROD_AccountList - Consumed read units: 1%
2013-05-21 07:33:05,479 - dynamic-dynamodb - INFO - PROD_AccountList - Consumed write units: 2%
2013-05-21 07:33:05,479 - dynamic-dynamodb - INFO - PROD_AccountList - Changing provisioning to 1300 read units and 33 write units
2013-05-21 07:33:05,506 - dynamic-dynamodb - WARNING - PROD_AccountList - Scaling limit exeeded. The table can only be scaled down twice per day.
2013-05-21 07:33:05,560 - dynamic-dynamodb - INFO - PROD_HashtagFanout - Consumed read units: 1%
2013-05-21 07:33:05,610 - dynamic-dynamodb - INFO - PROD_HashtagFanout - Consumed write units: 0%
2013-05-21 07:33:05,610 - dynamic-dynamodb - INFO - PROD_HashtagFanout - Scaling down writes is not done when usage is at 0%
2013-05-21 07:33:05,610 - dynamic-dynamodb - INFO - PROD_HashtagFanout - Changing provisioning to 650 read units and 100 write units
2013-05-21 07:33:05,637 - dynamic-dynamodb - WARNING - PROD_HashtagFanout - Scaling limit exeeded. The table can only be scaled down twice per day.
2013-05-21 07:33:05,697 - dynamic-dynamodb - INFO - PROD_PostrefsFeatured - Consumed read units: 1%
2013-05-21 07:33:05,789 - dynamic-dynamodb - INFO - PROD_PostrefsFeatured - Consumed write units: 0%
2013-05-21 07:33:05,789 - dynamic-dynamodb - INFO - PROD_PostrefsFeatured - Scaling down writes is not done when usage is at 0%
2013-05-21 07:33:05,789 - dynamic-dynamodb - INFO - PROD_PostrefsFeatured - Changing provisioning to 65 read units and 200 write units
2013-05-21 07:33:05,818 - dynamic-dynamodb - WARNING - PROD_PostrefsFeatured - Scaling limit exeeded. The table can only be scaled down twice per day.
2013-05-21 07:33:05,876 - dynamic-dynamodb - INFO - PROD_HashtagCounts - Consumed read units: 2%
2013-05-21 07:33:05,918 - dynamic-dynamodb - INFO - PROD_HashtagCounts - Consumed write units: 0%
2013-05-21 07:33:05,918 - dynamic-dynamodb - INFO - PROD_HashtagCounts - Scaling down writes is not done when usage is at 0%
2013-05-21 07:33:05,919 - dynamic-dynamodb - INFO - PROD_HashtagCounts - Changing provisioning to 33 read units and 50 write units
2013-05-21 07:33:05,947 - dynamic-dynamodb - WARNING - PROD_HashtagCounts - Scaling limit exeeded. The table can only be scaled down twice per day.
2013-05-21 07:33:06,021 - dynamic-dynamodb - INFO - PROD_HashtagSubscription - Consumed read units: 0%
2013-05-21 07:33:06,021 - dynamic-dynamodb - INFO - PROD_HashtagSubscription - Scaling down reads is not done when usage is at 0%
2013-05-21 07:33:06,083 - dynamic-dynamodb - INFO - PROD_HashtagSubscription - Consumed write units: 0%
2013-05-21 07:33:06,084 - dynamic-dynamodb - INFO - PROD_HashtagSubscription - Scaling down writes is not done when usage is at 0%
2013-05-21 07:33:06,084 - dynamic-dynamodb - INFO - PROD_HashtagSubscription - No need to change provisioning
2013-05-21 07:33:06,126 - dynamic-dynamodb - INFO - PROD_AccountCounts - Consumed read units: 8%
2013-05-21 07:33:06,171 - dynamic-dynamodb - INFO - PROD_AccountCounts - Consumed write units: 12%
2013-05-21 07:33:06,171 - dynamic-dynamodb - INFO - PROD_AccountCounts - Changing provisioning to 163 read units and 33 write units
2013-05-21 07:33:06,195 - dynamic-dynamodb - WARNING - PROD_AccountCounts - Scaling limit exeeded. The table can only be scaled down twice per day.
2013-05-21 07:33:06,277 - dynamic-dynamodb - INFO - PROD_PostrefsByUserWithoutReposts - Consumed read units: 1%
2013-05-21 07:33:06,323 - dynamic-dynamodb - INFO - PROD_PostrefsByUserWithoutReposts - Consumed write units: 1%
2013-05-21 07:33:06,324 - dynamic-dynamodb - INFO - PROD_PostrefsByUserWithoutReposts - Changing provisioning to 65 read units and 325 write units
2013-05-21 07:33:06,347 - dynamic-dynamodb - WARNING - PROD_PostrefsByUserWithoutReposts - Scaling limit exeeded. The table can only be scaled down twice per day.

The error message is wrong, I was able to size the CUs down manually in the AWS console (and am 100% sure that nobody had touched provisioning yet).

Config File:

# How often should Dynamic DynamoDB monitor changes (in seconds)
check-interval: 300

# Circuit breaker configuration
# No provisioning updates will be made unless this URL returns
# a HTTP 200 OK status code
#circuit-breaker-url: http://my.service.com/v1/is_up
#circuit-breaker-timeout: 500

[logging]
# Log level [debug|info|warning|error]
log-level: info
# Log file (comment out to get only console output)
log-file: /var/log/dynamic-dynamodb.log

# -------------------------------------------------------------------------------

[table: PROD_PostrefsForUser]
# -------------------------------------------------------------------------------
# Units to increase or decrease reads with, must be either percent or units
increase-reads-unit: percent
decrease-reads-unit: percent
increase-writes-unit: percent
decrease-writes-unit: percent

# Thresholds for scaling up or down the provisioning 
reads-upper-threshold: 90
reads-lower-threshold: 40
writes-upper-threshold: 90
writes-lower-threshold: 30

# How many percent should Dynamic DynamoDB increase/decrease provisioning with (%)
increase-reads-with: 30
decrease-reads-with: 35
increase-writes-with: 30
decrease-writes-with: 35

# Dynamic DynamoDB will not provision any more or less Provisioning than this
min-provisioned-reads: 300
max-provisioned-reads: 5000
min-provisioned-writes: 200
max-provisioned-writes: 3000



[table: PROD_AccountCounts]
# -------------------------------------------------------------------------------
increase-reads-unit: percent
decrease-reads-unit: percent
increase-writes-unit: percent
decrease-writes-unit: percent

# Thresholds for scaling up or down the provisioning 
reads-upper-threshold: 90
reads-lower-threshold: 40
writes-upper-threshold: 90
writes-lower-threshold: 30

# How many percent should Dynamic DynamoDB increase/decrease provisioning with (%)
increase-reads-with: 30
decrease-reads-with: 35
increase-writes-with: 30
decrease-writes-with: 35

# Dynamic DynamoDB will not provision any more or less Provisioning than this
min-provisioned-reads: 20
max-provisioned-reads: 5000
min-provisioned-writes: 10
max-provisioned-writes: 3000



[table: PROD_AccountList]              
# -------------------------------------------------------------------------------
increase-reads-unit: percent
decrease-reads-unit: percent
increase-writes-unit: percent
decrease-writes-unit: percent

# Thresholds for scaling up or down the provisioning 
reads-upper-threshold: 90
reads-lower-threshold: 40
writes-upper-threshold: 90
writes-lower-threshold: 30

# How many percent should Dynamic DynamoDB increase/decrease provisioning with (%)
increase-reads-with: 30
decrease-reads-with: 35
increase-writes-with: 30
decrease-writes-with: 35

# Dynamic DynamoDB will not provision any more or less Provisioning than this
min-provisioned-reads: 20
max-provisioned-reads: 5000
min-provisioned-writes: 10
max-provisioned-writes: 3000



[table: PROD_AccountRelation]
# -------------------------------------------------------------------------------
increase-reads-unit: percent
decrease-reads-unit: percent
increase-writes-unit: percent
decrease-writes-unit: percent

# Thresholds for scaling up or down the provisioning 
reads-upper-threshold: 90
reads-lower-threshold: 40
writes-upper-threshold: 90
writes-lower-threshold: 30

# How many percent should Dynamic DynamoDB increase/decrease provisioning with (%)
increase-reads-with: 30
decrease-reads-with: 35
increase-writes-with: 30
decrease-writes-with: 35

# Dynamic DynamoDB will not provision any more or less Provisioning than this
min-provisioned-reads: 20
max-provisioned-reads: 5000
min-provisioned-writes: 10
max-provisioned-writes: 3000



[table: PROD_HashIsAccount]
# -------------------------------------------------------------------------------
increase-reads-unit: percent
decrease-reads-unit: percent
increase-writes-unit: percent
decrease-writes-unit: percent

# Thresholds for scaling up or down the provisioning 
reads-upper-threshold: 90
reads-lower-threshold: 40
writes-upper-threshold: 90
writes-lower-threshold: 30

# How many percent should Dynamic DynamoDB increase/decrease provisioning with (%)
increase-reads-with: 30
decrease-reads-with: 35
increase-writes-with: 30
decrease-writes-with: 35

# Dynamic DynamoDB will not provision any more or less Provisioning than this
min-provisioned-reads: 20
max-provisioned-reads: 5000
min-provisioned-writes: 10
max-provisioned-writes: 3000



[table: PROD_HashKnownByAccount]
# -------------------------------------------------------------------------------
increase-reads-unit: percent
decrease-reads-unit: percent
increase-writes-unit: percent
decrease-writes-unit: percent

# Thresholds for scaling up or down the provisioning 
reads-upper-threshold: 90
reads-lower-threshold: 40
writes-upper-threshold: 90
writes-lower-threshold: 30

# How many percent should Dynamic DynamoDB increase/decrease provisioning with (%)
increase-reads-with: 30
decrease-reads-with: 35
increase-writes-with: 30
decrease-writes-with: 35

# Dynamic DynamoDB will not provision any more or less Provisioning than this
min-provisioned-reads: 20
max-provisioned-reads: 5000
min-provisioned-writes: 10
max-provisioned-writes: 3000



[table: PROD_HashtagCounts]         
# -------------------------------------------------------------------------------
increase-reads-unit: percent
decrease-reads-unit: percent
increase-writes-unit: percent
decrease-writes-unit: percent

# Thresholds for scaling up or down the provisioning 
reads-upper-threshold: 90
reads-lower-threshold: 40
writes-upper-threshold: 90
writes-lower-threshold: 30

# How many percent should Dynamic DynamoDB increase/decrease provisioning with (%)
increase-reads-with: 30
decrease-reads-with: 35
increase-writes-with: 30
decrease-writes-with: 35

# Dynamic DynamoDB will not provision any more or less Provisioning than this
min-provisioned-reads: 20
max-provisioned-reads: 5000
min-provisioned-writes: 10
max-provisioned-writes: 3000



[table: PROD_HashtagFanout]
# -------------------------------------------------------------------------------
increase-reads-unit: percent
decrease-reads-unit: percent
increase-writes-unit: percent
decrease-writes-unit: percent

# Thresholds for scaling up or down the provisioning 
reads-upper-threshold: 90
reads-lower-threshold: 40
writes-upper-threshold: 90
writes-lower-threshold: 30

# How many percent should Dynamic DynamoDB increase/decrease provisioning with (%)
increase-reads-with: 30
decrease-reads-with: 35
increase-writes-with: 30
decrease-writes-with: 35

# Dynamic DynamoDB will not provision any more or less Provisioning than this
min-provisioned-reads: 20
max-provisioned-reads: 5000
min-provisioned-writes: 10
max-provisioned-writes: 3000



[table: PROD_HashtagSubscription]
# -------------------------------------------------------------------------------
increase-reads-unit: percent
decrease-reads-unit: percent
increase-writes-unit: percent
decrease-writes-unit: percent

# Thresholds for scaling up or down the provisioning 
reads-upper-threshold: 90
reads-lower-threshold: 40
writes-upper-threshold: 90
writes-lower-threshold: 30

# How many percent should Dynamic DynamoDB increase/decrease provisioning with (%)
increase-reads-with: 30
decrease-reads-with: 35
increase-writes-with: 30
decrease-writes-with: 35

# Dynamic DynamoDB will not provision any more or less Provisioning than this
min-provisioned-reads: 20
max-provisioned-reads: 5000
min-provisioned-writes: 10
max-provisioned-writes: 3000



[table: PROD_OverallCounts]
# -------------------------------------------------------------------------------
increase-reads-unit: percent
decrease-reads-unit: percent
increase-writes-unit: percent
decrease-writes-unit: percent

# Thresholds for scaling up or down the provisioning 
reads-upper-threshold: 90
reads-lower-threshold: 40
writes-upper-threshold: 90
writes-lower-threshold: 30

# How many percent should Dynamic DynamoDB increase/decrease provisioning with (%)
increase-reads-with: 30
decrease-reads-with: 35
increase-writes-with: 30
decrease-writes-with: 35

# Dynamic DynamoDB will not provision any more or less Provisioning than this
min-provisioned-reads: 20
max-provisioned-reads: 5000
min-provisioned-writes: 10
max-provisioned-writes: 3000



[table: PROD_PostInfo]
# -------------------------------------------------------------------------------
increase-reads-unit: percent
decrease-reads-unit: percent
increase-writes-unit: percent
decrease-writes-unit: percent

# Thresholds for scaling up or down the provisioning 
reads-upper-threshold: 90
reads-lower-threshold: 40
writes-upper-threshold: 90
writes-lower-threshold: 30

# How many percent should Dynamic DynamoDB increase/decrease provisioning with (%)
increase-reads-with: 30
decrease-reads-with: 35
increase-writes-with: 30
decrease-writes-with: 35

# Dynamic DynamoDB will not provision any more or less Provisioning than this
min-provisioned-reads: 20
max-provisioned-reads: 5000
min-provisioned-writes: 10
max-provisioned-writes: 3000



[table: PROD_PostReaction]
# -------------------------------------------------------------------------------
increase-reads-unit: percent
decrease-reads-unit: percent
increase-writes-unit: percent
decrease-writes-unit: percent

# Thresholds for scaling up or down the provisioning 
reads-upper-threshold: 90
reads-lower-threshold: 40
writes-upper-threshold: 90
writes-lower-threshold: 30

# How many percent should Dynamic DynamoDB increase/decrease provisioning with (%)
increase-reads-with: 30
decrease-reads-with: 35
increase-writes-with: 30
decrease-writes-with: 35

# Dynamic DynamoDB will not provision any more or less Provisioning than this
min-provisioned-reads: 20
max-provisioned-reads: 5000
min-provisioned-writes: 10
max-provisioned-writes: 3000



[table: PROD_PostrefsByUserTextComments]
# -------------------------------------------------------------------------------
increase-reads-unit: percent
decrease-reads-unit: percent
increase-writes-unit: percent
decrease-writes-unit: percent

# Thresholds for scaling up or down the provisioning 
reads-upper-threshold: 90
reads-lower-threshold: 40
writes-upper-threshold: 90
writes-lower-threshold: 30

# How many percent should Dynamic DynamoDB increase/decrease provisioning with (%)
increase-reads-with: 30
decrease-reads-with: 35
increase-writes-with: 30
decrease-writes-with: 35

# Dynamic DynamoDB will not provision any more or less Provisioning than this
min-provisioned-reads: 20
max-provisioned-reads: 5000
min-provisioned-writes: 10
max-provisioned-writes: 3000



[table: PROD_PostrefsByUserWithReposts]
# -------------------------------------------------------------------------------
increase-reads-unit: percent
decrease-reads-unit: percent
increase-writes-unit: percent
decrease-writes-unit: percent

# Thresholds for scaling up or down the provisioning 
reads-upper-threshold: 90
reads-lower-threshold: 40
writes-upper-threshold: 90
writes-lower-threshold: 30

# How many percent should Dynamic DynamoDB increase/decrease provisioning with (%)
increase-reads-with: 30
decrease-reads-with: 35
increase-writes-with: 30
decrease-writes-with: 35

# Dynamic DynamoDB will not provision any more or less Provisioning than this
min-provisioned-reads: 20
max-provisioned-reads: 5000
min-provisioned-writes: 10
max-provisioned-writes: 3000



[table: PROD_PostrefsByUserWithoutReposts]
# -------------------------------------------------------------------------------
increase-reads-unit: percent
decrease-reads-unit: percent
increase-writes-unit: percent
decrease-writes-unit: percent

# Thresholds for scaling up or down the provisioning 
reads-upper-threshold: 90
reads-lower-threshold: 40
writes-upper-threshold: 90
writes-lower-threshold: 30

# How many percent should Dynamic DynamoDB increase/decrease provisioning with (%)
increase-reads-with: 30
decrease-reads-with: 35
increase-writes-with: 30
decrease-writes-with: 35

# Dynamic DynamoDB will not provision any more or less Provisioning than this
min-provisioned-reads: 20
max-provisioned-reads: 5000
min-provisioned-writes: 10
max-provisioned-writes: 3000



[table: PROD_PostrefsFeatured]
# -------------------------------------------------------------------------------
increase-reads-unit: percent
decrease-reads-unit: percent
increase-writes-unit: percent
decrease-writes-unit: percent

# Thresholds for scaling up or down the provisioning 
reads-upper-threshold: 90
reads-lower-threshold: 40
writes-upper-threshold: 90
writes-lower-threshold: 30

# How many percent should Dynamic DynamoDB increase/decrease provisioning with (%)
increase-reads-with: 30
decrease-reads-with: 35
increase-writes-with: 30
decrease-writes-with: 35

# Dynamic DynamoDB will not provision any more or less Provisioning than this
min-provisioned-reads: 20
max-provisioned-reads: 5000
min-provisioned-writes: 10
max-provisioned-writes: 3000



[table: PROD_PostrefsForUser]
# -------------------------------------------------------------------------------
increase-reads-unit: percent
decrease-reads-unit: percent
increase-writes-unit: percent
decrease-writes-unit: percent

# Thresholds for scaling up or down the provisioning 
reads-upper-threshold: 90
reads-lower-threshold: 40
writes-upper-threshold: 90
writes-lower-threshold: 30

# How many percent should Dynamic DynamoDB increase/decrease provisioning with (%)
increase-reads-with: 30
decrease-reads-with: 35
increase-writes-with: 30
decrease-writes-with: 35

# Dynamic DynamoDB will not provision any more or less Provisioning than this
min-provisioned-reads: 20
max-provisioned-reads: 5000
min-provisioned-writes: 10
max-provisioned-writes: 3000



[table: PROD_PostrefsPingPong]            
# -------------------------------------------------------------------------------
increase-reads-unit: percent
decrease-reads-unit: percent
increase-writes-unit: percent
decrease-writes-unit: percent

# Thresholds for scaling up or down the provisioning 
reads-upper-threshold: 90
reads-lower-threshold: 40
writes-upper-threshold: 90
writes-lower-threshold: 30

# How many percent should Dynamic DynamoDB increase/decrease provisioning with (%)
increase-reads-with: 30
decrease-reads-with: 35
increase-writes-with: 30
decrease-writes-with: 35

# Dynamic DynamoDB will not provision any more or less Provisioning than this
min-provisioned-reads: 20
max-provisioned-reads: 5000
min-provisioned-writes: 10
max-provisioned-writes: 3000



[table: PROD_PostrefsRepostLookup]
# -------------------------------------------------------------------------------
increase-reads-unit: percent
decrease-reads-unit: percent
increase-writes-unit: percent
decrease-writes-unit: percent

# Thresholds for scaling up or down the provisioning 
reads-upper-threshold: 90
reads-lower-threshold: 40
writes-upper-threshold: 90
writes-lower-threshold: 30

# How many percent should Dynamic DynamoDB increase/decrease provisioning with (%)
increase-reads-with: 30
decrease-reads-with: 35
increase-writes-with: 30
decrease-writes-with: 35

# Dynamic DynamoDB will not provision any more or less Provisioning than this
min-provisioned-reads: 20
max-provisioned-reads: 5000
min-provisioned-writes: 10
max-provisioned-writes: 3000



[table: PROD_PostrefsThread]
# -------------------------------------------------------------------------------
increase-reads-unit: percent
decrease-reads-unit: percent
increase-writes-unit: percent
decrease-writes-unit: percent

# Thresholds for scaling up or down the provisioning 
reads-upper-threshold: 90
reads-lower-threshold: 40
writes-upper-threshold: 90
writes-lower-threshold: 30

# How many percent should Dynamic DynamoDB increase/decrease provisioning with (%)
increase-reads-with: 30
decrease-reads-with: 35
increase-writes-with: 30
decrease-writes-with: 35

# Dynamic DynamoDB will not provision any more or less Provisioning than this
min-provisioned-reads: 20
max-provisioned-reads: 5000
min-provisioned-writes: 10
max-provisioned-writes: 3000



[table: PROD_Posts]
# -------------------------------------------------------------------------------
increase-reads-unit: percent
decrease-reads-unit: percent
increase-writes-unit: percent
decrease-writes-unit: percent

# Thresholds for scaling up or down the provisioning 
reads-upper-threshold: 90
reads-lower-threshold: 40
writes-upper-threshold: 90
writes-lower-threshold: 30

# How many percent should Dynamic DynamoDB increase/decrease provisioning with (%)
increase-reads-with: 30
decrease-reads-with: 35
increase-writes-with: 30
decrease-writes-with: 35

# Dynamic DynamoDB will not provision any more or less Provisioning than this
min-provisioned-reads: 20
max-provisioned-reads: 5000
min-provisioned-writes: 10
max-provisioned-writes: 3000



[table: PROD_ReactionCounts]
# -------------------------------------------------------------------------------
increase-reads-unit: percent
decrease-reads-unit: percent
increase-writes-unit: percent
decrease-writes-unit: percent

# Thresholds for scaling up or down the provisioning 
reads-upper-threshold: 90
reads-lower-threshold: 40
writes-upper-threshold: 90
writes-lower-threshold: 30

# How many percent should Dynamic DynamoDB increase/decrease provisioning with (%)
increase-reads-with: 30
decrease-reads-with: 35
increase-writes-with: 30
decrease-writes-with: 35

# Dynamic DynamoDB will not provision any more or less Provisioning than this
min-provisioned-reads: 20
max-provisioned-reads: 5000
min-provisioned-writes: 10
max-provisioned-writes: 3000



[table: PROD_ReactionUserAbuseReport]
# -------------------------------------------------------------------------------
increase-reads-unit: percent
decrease-reads-unit: percent
increase-writes-unit: percent
decrease-writes-unit: percent

# Thresholds for scaling up or down the provisioning 
reads-upper-threshold: 90
reads-lower-threshold: 40
writes-upper-threshold: 90
writes-lower-threshold: 30

# How many percent should Dynamic DynamoDB increase/decrease provisioning with (%)
increase-reads-with: 30
decrease-reads-with: 35
increase-writes-with: 30
decrease-writes-with: 35

# Dynamic DynamoDB will not provision any more or less Provisioning than this
min-provisioned-reads: 20
max-provisioned-reads: 5000
min-provisioned-writes: 10
max-provisioned-writes: 3000



[table: PROD_ReactionUserDislike]
# -------------------------------------------------------------------------------
increase-reads-unit: percent
decrease-reads-unit: percent
increase-writes-unit: percent
decrease-writes-unit: percent

# Thresholds for scaling up or down the provisioning 
reads-upper-threshold: 90
reads-lower-threshold: 40
writes-upper-threshold: 90
writes-lower-threshold: 30

# How many percent should Dynamic DynamoDB increase/decrease provisioning with (%)
increase-reads-with: 30
decrease-reads-with: 35
increase-writes-with: 30
decrease-writes-with: 35

# Dynamic DynamoDB will not provision any more or less Provisioning than this
min-provisioned-reads: 20
max-provisioned-reads: 5000
min-provisioned-writes: 10
max-provisioned-writes: 3000



[table: PROD_ReactionUserLike]
# -------------------------------------------------------------------------------
increase-reads-unit: percent
decrease-reads-unit: percent
increase-writes-unit: percent
decrease-writes-unit: percent

# Thresholds for scaling up or down the provisioning 
reads-upper-threshold: 90
reads-lower-threshold: 40
writes-upper-threshold: 90
writes-lower-threshold: 30

# How many percent should Dynamic DynamoDB increase/decrease provisioning with (%)
increase-reads-with: 30
decrease-reads-with: 35
increase-writes-with: 30
decrease-writes-with: 35

# Dynamic DynamoDB will not provision any more or less Provisioning than this
min-provisioned-reads: 20
max-provisioned-reads: 5000
min-provisioned-writes: 10
max-provisioned-writes: 3000



[table: PROD_ReactionUserRepost]
# -------------------------------------------------------------------------------
increase-reads-unit: percent
decrease-reads-unit: percent
increase-writes-unit: percent
decrease-writes-unit: percent

# Thresholds for scaling up or down the provisioning 
reads-upper-threshold: 90
reads-lower-threshold: 40
writes-upper-threshold: 90
writes-lower-threshold: 30

# How many percent should Dynamic DynamoDB increase/decrease provisioning with (%)
increase-reads-with: 30
decrease-reads-with: 35
increase-writes-with: 30
decrease-writes-with: 35

# Dynamic DynamoDB will not provision any more or less Provisioning than this
min-provisioned-reads: 20
max-provisioned-reads: 5000
min-provisioned-writes: 10
max-provisioned-writes: 3000



[table: PROD_RescrapeSpool]
# -------------------------------------------------------------------------------
increase-reads-unit: percent
decrease-reads-unit: percent
increase-writes-unit: percent
decrease-writes-unit: percent

# Thresholds for scaling up or down the provisioning 
reads-upper-threshold: 90
reads-lower-threshold: 40
writes-upper-threshold: 90
writes-lower-threshold: 30

# How many percent should Dynamic DynamoDB increase/decrease provisioning with (%)
increase-reads-with: 30
decrease-reads-with: 35
increase-writes-with: 30
decrease-writes-with: 35

# Dynamic DynamoDB will not provision any more or less Provisioning than this
min-provisioned-reads: 20
max-provisioned-reads: 5000
min-provisioned-writes: 10
max-provisioned-writes: 3000



# Maintenance windows (in UTC)
#maintenance-windows: 22:00-23:59,00:00-06:00

# Other settings

# Allow down scaling when at 0% consumed reads
#allow-scaling-down-reads-on-0-percent: true
#allow-scaling-down-writes-on-0-percent: true

# Restric scale down to only happend when BOTH reads AND writes are in need
# of scaling down. Set this to "true" to minimize down scaling.
#always-decrease-rw-together: true

Handle ResourceInUseException better

Handle ResourceInUseException better. This exception is thrown when down scaling of a table is in progress when the next down scaling is performed.

2013-04-26 08:53:00,837 - dynamic-dynamodb - INFO - prod-displays-20130426 - Consumed read units: 20%
2013-04-26 08:53:00,898 - dynamic-dynamodb - INFO - prod-displays-20130426 - Consumed write units: 54%
2013-04-26 08:53:00,898 - dynamic-dynamodb - INFO - prod-displays-20130426 - Changing provisioning to 25 read units and 500 write units
2013-04-26 08:53:00,905 - dynamic-dynamodb - WARNING - prod-displays-20130426 - Not performing throughput changes when table is in UPDATING state
2013-04-26 08:53:00,913 - dynamic-dynamodb - ERROR - prod-displays-20130426 - Unhandled exception: ResourceInUseException: Attempt to change a resource which is still in use: Table is being updated: prod-displays-20130426. Please file a bug report at https://github.com/sebdah/dynamic-dynamodb/issues

"only downscale reads AND writes" option would be useful

It would be awesome if one could "queue" a table's provisioning decreases until decreases for both reads AND writes would be due.

Background: One can only send two downscaling requests per day -- not two for reads and two for writes. Optionally downscaling reads and writes simultaneously would increase efficiency of using this limitation.

Enhance configuration option validation

Enhance configuration option validation so that e.g. it is not possible to increase writes with more than 100% for example.

Allow scale down at 0% consumed count

Dynamic DynamoDB is only scaling down when consumed capacity > 0%. That has been implemented as a security enhancement, but in some cases (where the table really has low traffic) it makes sense to scale down at 0%.

Create an option to make 0% down scaling possible.

dynamic-dynamodb --version causes AttributeError in cli

dynamic-dynamodb --version

Calling the above from a terminal results in:

Traceback (most recent call last):
  File "/usr/local/bin/dynamic-dynamodb", line 22, in <module>
    import dynamic_dynamodb
  File "/usr/local/lib/python2.7/dist-packages/dynamic_dynamodb/__init__.py", line 26, in <module>
    import core
  File "/usr/local/lib/python2.7/dist-packages/dynamic_dynamodb/core/__init__.py", line 6, in <module>
    import dynamodb
  File "/usr/local/lib/python2.7/dist-packages/dynamic_dynamodb/core/dynamodb.py", line 7, in <module>
    from dynamic_dynamodb.log_handler import LOGGER as logger
  File "/usr/local/lib/python2.7/dist-packages/dynamic_dynamodb/log_handler.py", line 23, in <module>
    import config_handler
  File "/usr/local/lib/python2.7/dist-packages/dynamic_dynamodb/config_handler.py", line 4, in <module>
    CONFIGURATION = config.get_configuration()
  File "/usr/local/lib/python2.7/dist-packages/dynamic_dynamodb/config/__init__.py", line 63, in get_configuration
    cmd_line_options = command_line_parser.parse()
  File "/usr/local/lib/python2.7/dist-packages/dynamic_dynamodb/config/command_line_parser.py", line 114, in parse
    print 'Dynamic DynamoDB version: {0}'.format(dynamic_dynamodb.version())
AttributeError: 'module' object has no attribute 'version'

Not a terrible huge issue, just a little inconvenient...

Improper scaling under certain circumstances

Improper scaling under certain circumstances results in bad read/write provisioning.

Increase in percent does not add to current provisioning properly

Unhandled exception: ValidationException

ubuntu@helo:~$ sudo dynamic-dynamodb -c PROD_AccountCounts.conf
2013-04-11 07:34:45,617 - dynamic-dynamodb - INFO - Consumed reads: 34
2013-04-11 07:34:45,630 - dynamic-dynamodb - INFO - Provisioned reads: 200
2013-04-11 07:34:45,630 - dynamic-dynamodb - INFO - Consumed read capacity: 17%
2013-04-11 07:34:45,630 - dynamic-dynamodb - INFO - Reads lower threshold: 30%
2013-04-11 07:34:45,630 - dynamic-dynamodb - INFO - Reads upper threshold: 80%
2013-04-11 07:34:45,666 - dynamic-dynamodb - INFO - Consumed writes: 7
2013-04-11 07:34:45,680 - dynamic-dynamodb - INFO - Provisioned writes: 18
2013-04-11 07:34:45,680 - dynamic-dynamodb - INFO - Consumed write capacity: 39%
2013-04-11 07:34:45,681 - dynamic-dynamodb - INFO - Writes lower threshold: 30%
2013-04-11 07:34:45,681 - dynamic-dynamodb - INFO - Writes upper threshold: 95%
2013-04-11 07:34:45,720 - dynamic-dynamodb - INFO - Reached provisioned reads min limit: 200
2013-04-11 07:34:45,720 - dynamic-dynamodb - INFO - Changing provisioning to 200 reads and 18 writes
Traceback (most recent call last):
  File "/usr/local/bin/dynamic-dynamodb", line 7, in <module>
    main.main()
  File "/usr/local/lib/python2.6/dist-packages/dynamic_dynamodb/main.py", line 229, in main
    dynamic_ddb.run()
  File "/usr/local/lib/python2.6/dist-packages/dynamic_dynamodb/dynamic_dynamodb.py", line 146, in run
    self.ensure_proper_provisioning()
  File "/usr/local/lib/python2.6/dist-packages/dynamic_dynamodb/dynamic_dynamodb.py", line 210, in ensure_proper_provisioning
    throughput['write_units'])
  File "/usr/local/lib/python2.6/dist-packages/dynamic_dynamodb/dynamic_dynamodb.py", line 483, in _update_throughput
    table.update_throughput(int(read_units), int(write_units))
  File "/usr/local/lib/python2.6/dist-packages/boto/dynamodb/table.py", line 245, in update_throughput
    self.layer2.update_throughput(self, read_units, write_units)
  File "/usr/local/lib/python2.6/dist-packages/boto/dynamodb/layer2.py", line 405, in update_throughput
    'WriteCapacityUnits': write_units})
  File "/usr/local/lib/python2.6/dist-packages/boto/dynamodb/layer1.py", line 261, in update_table
    return self.make_request('UpdateTable', json_input)
  File "/usr/local/lib/python2.6/dist-packages/boto/dynamodb/layer1.py", line 118, in make_request
    retry_handler=self._retry_handler)
  File "/usr/local/lib/python2.6/dist-packages/boto/connection.py", line 845, in _mexe
    status = retry_handler(response, i, next_sleep)
  File "/usr/local/lib/python2.6/dist-packages/boto/dynamodb/layer1.py", line 155, in _retry_handler
    response.status, response.reason, data)
boto.dynamodb.exceptions.DynamoDBValidationError: DynamoDBValidationError: 400 Bad Request
{'message': 'The provisioned throughput for the table will not change. The requested value equals the current value. Current ReadCapacityUnits provisioned for the table: 200. Requested ReadCapacityUnits: 200. Current WriteCapacityUnits provisioned for the table: 18. Requested WriteCapacityUnits: 18. Refer to the Amazon DynamoDB Developer Guide for current limits and how to request higher limits.', '__type': 'com.amazon.coral.validate#ValidationException'}
ubuntu@helo:~$

Update --help docs

Update --help docs

Restructure the Dynamic DynamoDB code base

Fix Python 2.6 support

This may include many things, but at least all formatting tags must be numbered.

No handler was ready to authenticate

I'm getting this error once per every 100-200 runs of dynamic-dynamo (approximately):

2013-08-13 18:17:13,792 - dynamic-dynamodb - ERROR - Failed connecting to DynamoDB: No handler was ready to authenticate. 1 handlers were checked. ['HmacAuthV4Handler'] Check your credentials
:  2013-08-13 18:17:13,792 - dynamic-dynamodb - ERROR - Please report an issue at: https://github.com/sebdah/dynamic-dynamodb/issues
:  Traceback (most recent call last):
:    File "/usr/local/bin/dynamic-dynamodb", line 22, in <module>
:      import dynamic_dynamodb
:    File "/usr/local/lib/python2.7/dist-packages/dynamic_dynamodb/__init__.py", line 26, in <module>
:      import core
:    File "/usr/local/lib/python2.7/dist-packages/dynamic_dynamodb/core/__init__.py", line 6, in <module>
:      import dynamodb
:    File "/usr/local/lib/python2.7/dist-packages/dynamic_dynamodb/core/dynamodb.py", line 56, in <module>
:      DYNAMODB_CONNECTION = __get_connection_dynamodb()
:    File "/usr/local/lib/python2.7/dist-packages/dynamic_dynamodb/core/dynamodb.py", line 23, in __get_connection_dynamodb
:      configuration['global']['region'])
:    File "/usr/local/lib/python2.7/dist-packages/boto/dynamodb/__init__.py", line 65, in connect_to_region
:      return region.connect(**kw_params)
:    File "/usr/local/lib/python2.7/dist-packages/boto/regioninfo.py", line 63, in connect
:      return self.connection_cls(region=self, **kw_params)
:    File "/usr/local/lib/python2.7/dist-packages/boto/dynamodb/layer2.py", line 152, in __init__
:      validate_certs=validate_certs)
:    File "/usr/local/lib/python2.7/dist-packages/boto/dynamodb/layer1.py", line 92, in __init__
:      validate_certs=validate_certs)
:    File "/usr/local/lib/python2.7/dist-packages/boto/connection.py", line 554, in __init__
:      host, config, self.provider, self._required_auth_capability())
:    File "/usr/local/lib/python2.7/dist-packages/boto/auth.py", line 686, in get_auth_handler
:      'Check your credentials' % (len(names), str(names)))
:  boto.exception.NoAuthHandlerFound: No handler was ready to authenticate. 1 handlers were checked. ['HmacAuthV4Handler'] Check your credentials

Handle missing table exceptions

Stack trace:

Traceback (most recent call last):
  File "./dynamic-dynamodb", line 7, in <module>
    main.main()
  File "/Users/sebastian/git/dynamic-dynamodb/dynamic_dynamodb/main.py", line 174, in main
    dynamic_ddb.run()
  File "/Users/sebastian/git/dynamic-dynamodb/dynamic_dynamodb/dynamic_dynamodb.py", line 148, in run
    self.ensure_proper_provisioning()
  File "/Users/sebastian/git/dynamic-dynamodb/dynamic_dynamodb/dynamic_dynamodb.py", line 160, in ensure_proper_provisioning
    read_usage_percent = self._get_consumed_reads_percentage()
  File "/Users/sebastian/git/dynamic-dynamodb/dynamic_dynamodb/dynamic_dynamodb.py", line 322, in _get_consumed_reads_percentage
    float(consumed_reads) / float(self._get_provisioned_read_units()) * 100))
  File "/Users/sebastian/git/dynamic-dynamodb/dynamic_dynamodb/dynamic_dynamodb.py", line 365, in _get_provisioned_read_units
    table = self.ddb_connection.get_table(self.table_name)
  File "/usr/local/lib/python2.7/site-packages/boto/dynamodb/layer2.py", line 360, in get_table
    response = self.layer1.describe_table(name)
  File "/usr/local/lib/python2.7/site-packages/boto/dynamodb/layer1.py", line 217, in describe_table
    return self.make_request('DescribeTable', json_input)
  File "/usr/local/lib/python2.7/site-packages/boto/dynamodb/layer1.py", line 118, in make_request
    retry_handler=self._retry_handler)
  File "/usr/local/lib/python2.7/site-packages/boto/connection.py", line 845, in _mexe
    status = retry_handler(response, i, next_sleep)
  File "/usr/local/lib/python2.7/site-packages/boto/dynamodb/layer1.py", line 158, in _retry_handler
    data)
boto.exception.DynamoDBResponseError: DynamoDBResponseError: 400 Bad Request
{u'message': u'Requested resource not found: Table: prod-sssions-201304 not found', u'__type': u'com.amazonaws.dynamodb.v20111205#ResourceNotFoundException'}

Add --log-level option to command line

Add --log-level option to command line

Mixup of read and writes provisioing in scaling

Enhanced logging options

It should be possible to configure logging more in detail. E.g. logging to file should be possible as well as log level etc.

Allow to run multiple instances in parallel

Hi,
currently when dynamic-dynamodb starts in daemon mode, it creates a pid file on this location:
daemon = DynamicDynamoDBDaemon('/tmp/dynamic-dynamodb.pid')

Since this is harcoded, it is not possible to run multiple instances od dynamic-dynamodb.

Regards,

int() argument must be a string or a number, not 'NoneType'

I'm getting this exception:

2013-07-21 10:12:28,421 - dynamic-dynamodb - INFO - foo - Consumed read units: 0%
2013-07-21 10:12:28,427 - dynamic-dynamodb - INFO - foo - Scaling down reads is not done when usage is at 0%
2013-07-21 10:12:28,484 - dynamic-dynamodb - INFO - foo - Consumed write units: 40%
Jul 21 12:12:28 conveyer.rultor.com rultor:  Traceback (most recent call last):
Jul 21 12:12:28 conveyer.rultor.com rultor:    File "/usr/local/bin/dynamic-dynamodb", line 26, in <module>
Jul 21 12:12:28 conveyer.rultor.com rultor:      dynamic_dynamodb.main()
Jul 21 12:12:28 conveyer.rultor.com rultor:    File "/usr/local/lib/python2.7/dist-packages/dynamic_dynamodb/__init__.py", line 68, in main
Jul 21 12:12:28 conveyer.rultor.com rultor:      core.ensure_provisioning(table_name)
Jul 21 12:12:28 conveyer.rultor.com rultor:    File "/usr/local/lib/python2.7/dist-packages/dynamic_dynamodb/core/__init__.py", line 30, in ensure_provisioning
Jul 21 12:12:28 conveyer.rultor.com rultor:      table_name)
Jul 21 12:12:28 conveyer.rultor.com rultor:    File "/usr/local/lib/python2.7/dist-packages/dynamic_dynamodb/core/__init__.py", line 214, in __ensure_provisioning_writes
Jul 21 12:12:28 conveyer.rultor.com rultor:      get_table_option(table_name, 'decrease_writes_with'))
Jul 21 12:12:28 conveyer.rultor.com rultor:    File "/usr/local/lib/python2.7/dist-packages/dynamic_dynamodb/core/calculators.py", line 113, in decrease_writes_in_percent
Jul 21 12:12:28 conveyer.rultor.com rultor:      current_provisioning)
Jul 21 12:12:28 conveyer.rultor.com rultor:    File "/usr/local/lib/python2.7/dist-packages/dynamic_dynamodb/core/calculators.py", line 31, in get_min_provisioned_writes
Jul 21 12:12:28 conveyer.rultor.com rultor:      (current_provisioning * 2)))
Jul 21 12:12:28 conveyer.rultor.com rultor:  TypeError: int() argument must be a string or a number, not 'NoneType'

This is how I call the tool:

dynamic-dynamodb \
            --table-name foo \
            --reads-upper-threshold 90 \
            --reads-lower-threshold 30 \
            --increase-reads-with 50 \
            --decrease-reads-with 50 \
            --writes-upper-threshold 90 \
            --writes-lower-threshold 40 \
            --increase-writes-with 50 \
            --decrease-writes-with 70

set cloudwatch alerts and listen on sqs queue for alerts, then take action

It would be nice to make this more dynamic and listen on an sqs queue for alerts and proactively react to them, instead of polling cloudwatch...
I can have a go at this, but cant promise it will be very elegant ;)

No upscaling happening when scaling limit is exceeded

If downscaling provisioning fails since two changes where made on the given day already, upscaling isn't executed either.

Example:

Table has 100 Read-CUs and 100 Write-CUs
Table has been scaled down twice already
dynamic-dynamo figures out the best provisioning would be 80 Read-CU and 200 Write-CU

Result: The Bump to 200 Write-CUs is not happening since the downscaling of the Read-CUs failed.
Expected behavior: The increase in provisioned capacity should be executed nevertheless.

This is dangerous bug, since dyn-dyn stops scaling up again even though upscaling is ALWAYS possible.

ValueError: Unknown format code 'd' for object of type 'str'

It seems that dynamic-dynamo stumbles over certain usage ratios.
This only happens with certain tables, not sure what makes the difference.

Example output:


ubuntu@blah:~$ dynamic-dynamodb --dry-run -c PROD_HashKnownByAccount.conf 
2013-04-04 14:38:01,342 - dynamic-dynamodb - dryrun - INFO - Consumed reads: 811
2013-04-04 14:38:01,342 - dynamic-dynamodb - dryrun - INFO - Provisioned reads: 1000
2013-04-04 14:38:01,342 - dynamic-dynamodb - dryrun - INFO - Read usage: 82%
2013-04-04 14:38:01,385 - dynamic-dynamodb - dryrun - INFO - Consumed writes: 15
2013-04-04 14:38:01,386 - dynamic-dynamodb - dryrun - INFO - Provisioned writes: 200
2013-04-04 14:38:01,386 - dynamic-dynamodb - dryrun - INFO - Write usage: 8%
Traceback (most recent call last):
  File "/usr/local/bin/dynamic-dynamodb", line 7, in <module>
    main.main()
  File "/usr/local/lib/python2.6/dist-packages/dynamic_dynamodb/main.py", line 174, in main
    dynamic_ddb.run()
  File "/usr/local/lib/python2.6/dist-packages/dynamic_dynamodb/dynamic_dynamodb.py", line 148, in run
    self.ensure_proper_provisioning()
  File "/usr/local/lib/python2.6/dist-packages/dynamic_dynamodb/dynamic_dynamodb.py", line 207, in ensure_proper_provisioning
    throughput['write_units']))
ValueError: Unknown format code 'd' for object of type 'str'

with this config:


# AWS region to use
region: us-east-1

# How often should Dynamic DynamoDB monitor changes (in seconds)
check-interval: 300

# -------------------------------------------------------------------------------
[table: PROD_HashKnownByAccount]
# -------------------------------------------------------------------------------
#
# Read provisioning configuration
#
# Thresholds for scaling up or down the provisioning (%)
reads-upper-threshold: 100
reads-lower-threshold: 30

# How many percent should Dynamic DynamoDB increase/decrease provisioning with (%)
increase-reads-with: 20
decrease-reads-with: 80

# Maximum and minimum read provisioning
# Dynamic DynamoDB will not provision any more or less reads than this
min-provisioned-reads: 10
max-provisioned-reads: 1500

#
# Write provisioning configuration
#
# Thresholds for scaling up or down the provisioning (%)
writes-upper-threshold: 100
writes-lower-threshold: 30

# How many percent should Dynamic DynamoDB increase/decrease provisioning with (%)
increase-writes-with: 30
decrease-writes-with: 60

# Maximum and minimum write provisioning
# Dynamic DynamoDB will not provision any more or less writes than this
min-provisioned-writes: 10
max-provisioned-writes: 500

Add support for max read/write provisioning

Can't you only scale up or down up to twice per day?

I believe you can only adjust your DynamoDB throughput up to twice per day, so why the need for this library?

Support for maintenance windows

Add support for maintenance windows. If specified Dynamic DynamoDB should only scale up/down during this window.

Multiple maintenance windows should be possible to add.

increase_reads_in_percent calculations are incorrect.

On line 46 in calculators.py in the function increase_reads_in_percent, you have

increase = int(
    float(current_provisioning)*(float(percent)/100+1))
updated_provisioning = current_provisioning + increase

it should just be

increase = int(float(current_provisioning)*(float(percent)/100))
updated_provisioning = current_provisioning + increase

or alternatively

updated_provisioning = int(float(current_provisioning)*(float(percent)/100+1))

"allow-scaling-down-reads-on-0-percent: true" is ignored

Even if "allow-scaling-down-reads-on-0-percent: true" is in the config file, dyndyn does not do so:

2013-05-17 23:35:31,866 - dynamic-dynamodb - INFO - PROD_PostrefsByUserTextComments - Scaling down reads is not done when usage is at 0%

logging through syslog

Would be nice to have an ability to log through syslog, in daemon mode, e.g.:

dynamic-dynamodb --instance foo --syslog dyndyn \
  --table-name foo --check-interval 300 --daemon start \

Where dyndyn is a facility for syslog.

Add support for dead-man's-switch API call

Any service can go down for whatever reason. In this case, DynamoDB Tables might no longer get any traffic. If an outage lasts long enough, provisioning might scale down to the min-provisioned-values. Since it makes sense to increase provisioning in smaller steps (since upscale-requests are not limited), it might take a long time after such an outage to get the table provisioning back to the required "normal" level.

Idea: Allow configuring a URL and max-response-time that is checked every check-interval before the dynamic-dynamo does its magic.

This might be a simple api endpoint that returns 200 if the service is up. If it responds, the script does its magic. If it does not return 200 or responds too slow, dynamic-dynamodb should stop downscaling the provisioning to allow speedy retrun to production throughput after an outage.

An attempt to update provisioning is made even if the requested values are equal to the tables current values

If "always-decrease-rw-together" is set to true, there is a chance for the requested provisioning change to be set to that of the current tables. Because of that among other reasons... there are times when a request is made to change a tables provisioning to the same values it is currently provisioned.

In the "update_throughput" function in "dynamic-dynamodb/core/init.py" I would recommend a simple check that determines if the requested provisioning values are both equal to the current provisioning values.

This snippet would do nicely around line 305-306 in the aforementioned file:

if read_units == table.read_units && write_units == table.write_units:
    return

Configure AWS credentials on command line

Make it possible to configure AWS credentials on command line

Handle DynamoDB provisioning exceptions cleaner

Update config file docs

Update provisioning in units not just percentage

For some table's access patterns, scaling the table in percent does not make sense and scaling by fixed amounts would make more sense.

Growth of table traffic isn't always exponential but can also be linear.

Add support for configuration files

Add support for regular Python configuration files. Here's a suggested configuration:

[global]
#aws-access-key-id: AWS_ACCESS_KEY
#aws-secret-access-key-id: AWS_SECRET_KEY
region: us-east-1
check-interval: 300

[table: my-table]
# Read provisioning configuration
reads-upper-threshold: 90
reads-lower-threshold: 30
increase-reads-with: 50
decrease-reads-with: 50
#min-provisioned-reads: 100
#max-provisioned-reads: 500

# Write provisioning configuration
writes-upper-threshold: 90
writes-lower-threshold: 30
increase-writes-with: 50
decrease-writes-with: 50
#min-provisioned-writes: 100
#max-provisioned-writes: 500

Handle multiple tables at once

It would be fantastic if Dynamic DynamoDB could handle more than one table at a time. Here's a suggested configuration:

[global]
#aws-access-key-id: AWS_ACCESS_KEY
#aws-secret-access-key-id: AWS_SECRET_KEY
region: us-east-1
check-interval: 300

[table: my-table]
# Read provisioning configuration
reads-upper-threshold: 90
reads-lower-threshold: 30
increase-reads-with: 50
decrease-reads-with: 50
#min-provisioned-reads: 100
#max-provisioned-reads: 500

# Write provisioning configuration
writes-upper-threshold: 90
writes-lower-threshold: 30
increase-writes-with: 50
decrease-writes-with: 50
#min-provisioned-writes: 100
#max-provisioned-writes: 500

It would then be possible to define more than 1 [table: x] section and Dynamic DynamoDB could iterate over those sections.

Min/max limits seems to be read improperly from configuration files

Add --log-file command line option

Add --log-file command line option in order to support writing to log file when using only command line options.

Recommend Projects

React

A declarative, efficient, and flexible JavaScript library for building user interfaces.
Vue.js

🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
Typescript

TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
TensorFlow

An Open Source Machine Learning Framework for Everyone
Django

The Web framework for perfectionists with deadlines.
Laravel

A PHP framework for web artisans
D3

Bring data to life with SVG, Canvas and HTML. 📊📈🎉

Recommend Topics

javascript

JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
web

Some thing interesting about web. New door for the world.
server

A server is a program made to process requests and deliver data to clients.
Machine learning

Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
Visualization

Some thing interesting about visualization, use data art
Game

Some thing interesting about game, make everyone happy.

Recommend Org

Facebook

We are working to build community through open source technology. NB: members must have two-factor auth.
Microsoft

Open source projects and samples from Microsoft.
Google

Google ❤️ Open Source for everyone.
Alibaba

Alibaba Open Source for everyone
D3

Data-Driven Documents codes.
Tencent

China tencent open source team.