[jira] [Commented] (MESOS-3890) Add notion of evictable task to RunTaskMessage
[
https://issues.apache.org/jira/browse/MESOS-3890?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15080370#comment-15080370
]
Guangya Liu commented on MESOS-3890:
[~hartem] [~jvanremoortere] [~kaysoky] Since the master cannot get the
evictable executors when launch task, the current implementation is we leave
this to agent to kill executors, shall we close this one directly? Thanks.
> Add notion of evictable task to RunTaskMessage
> --
>
> Key: MESOS-3890
> URL: https://issues.apache.org/jira/browse/MESOS-3890
> Project: Mesos
> Issue Type: Bug
>Reporter: Artem Harutyunyan
>Assignee: Guangya Liu
> Labels: mesosphere
>
> {code}
> message RunTaskMessage {
> ...
> // This list can be non-empty when a task is launched on reserved
> // resources. If the reserved resources are in use (as revocable
> // resources), this list contains the executors that can be evicted
> // to make room to run this task.
> repeated ExecutorID evictable_executors = 5;
> ...
> }
> {code}
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
[jira] [Commented] (MESOS-3897) Identify and implement test cases for verifying eviction logic in the allocator.
[ https://issues.apache.org/jira/browse/MESOS-3897?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15080371#comment-15080371 ] Guangya Liu commented on MESOS-3897: The eviction logic will be only in agent, [~hartem], do we still need this ticket? Thanks. > Identify and implement test cases for verifying eviction logic in the > allocator. > > > Key: MESOS-3897 > URL: https://issues.apache.org/jira/browse/MESOS-3897 > Project: Mesos > Issue Type: Bug >Reporter: Artem Harutyunyan >Assignee: Artem Harutyunyan > Labels: mesosphere > -- This message was sent by Atlassian JIRA (v6.3.4#6332)
[jira] [Comment Edited] (MESOS-3955) Add helper function to get stateless resources.
[
https://issues.apache.org/jira/browse/MESOS-3955?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15023812#comment-15023812
]
Guangya Liu edited comment on MESOS-3955 at 1/3/16 9:13 AM:
https://reviews.apache.org/r/40529/
was (Author: gyliu):
https://reviews.apache.org/r/40524/
https://reviews.apache.org/r/40529/
> Add helper function to get stateless resources.
> ---
>
> Key: MESOS-3955
> URL: https://issues.apache.org/jira/browse/MESOS-3955
> Project: Mesos
> Issue Type: Bug
>Reporter: Guangya Liu
>Assignee: Guangya Liu
>
> Stateless are {{No Persistent Volume}} resources, it’s used by allocator to
> calculate optimised resources (reserved().stateless() -allocated()).
> {code}
> class Resources {
> ...
> // Tests if the given Resource object has no stateful elements.
> static bool isStateless(const Resource& resource);
> ...
> // Returns the resources that do not have stateful elements.
> Resources stateless() const;
> ...
> }
> {code}
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
[jira] [Updated] (MESOS-4145) Update allocator to get allocation slack resources
[ https://issues.apache.org/jira/browse/MESOS-4145?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Guangya Liu updated MESOS-4145: --- Description: The allocator should be updated to 1) Report allocation slack resources when add a new agent. 2) Allocate allocation slack resources when sending offers. 3) Update allocation slack resources when update agent, update dynamic reservations with both framework and http endpoint. was:The allocator should be updated to report allocation slack resources to be used by launch tasks. > Update allocator to get allocation slack resources > -- > > Key: MESOS-4145 > URL: https://issues.apache.org/jira/browse/MESOS-4145 > Project: Mesos > Issue Type: Bug >Reporter: Guangya Liu >Assignee: Guangya Liu > > The allocator should be updated to > 1) Report allocation slack resources when add a new agent. > 2) Allocate allocation slack resources when sending offers. > 3) Update allocation slack resources when update agent, update dynamic > reservations with both framework and http endpoint. -- This message was sent by Atlassian JIRA (v6.3.4#6332)
[jira] [Commented] (MESOS-3890) Add notion of evictable task to RunTaskMessage
[
https://issues.apache.org/jira/browse/MESOS-3890?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15080372#comment-15080372
]
Klaus Ma commented on MESOS-3890:
-
Let's keep this open; we need this JIRA after post MVP (move CLI executor from
slave to master).
> Add notion of evictable task to RunTaskMessage
> --
>
> Key: MESOS-3890
> URL: https://issues.apache.org/jira/browse/MESOS-3890
> Project: Mesos
> Issue Type: Bug
>Reporter: Artem Harutyunyan
>Assignee: Guangya Liu
> Labels: mesosphere
>
> {code}
> message RunTaskMessage {
> ...
> // This list can be non-empty when a task is launched on reserved
> // resources. If the reserved resources are in use (as revocable
> // resources), this list contains the executors that can be evicted
> // to make room to run this task.
> repeated ExecutorID evictable_executors = 5;
> ...
> }
> {code}
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
[jira] [Comment Edited] (MESOS-4145) Update allocator to get allocation slack resources
[ https://issues.apache.org/jira/browse/MESOS-4145?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15055223#comment-15055223 ] Guangya Liu edited comment on MESOS-4145 at 1/3/16 9:20 AM: https://reviews.apache.org/r/40632/ https://reviews.apache.org/r/41847/ https://reviews.apache.org/r/41791/ was (Author: gyliu): https://reviews.apache.org/r/40632/ > Update allocator to get allocation slack resources > -- > > Key: MESOS-4145 > URL: https://issues.apache.org/jira/browse/MESOS-4145 > Project: Mesos > Issue Type: Bug >Reporter: Guangya Liu >Assignee: Guangya Liu > > The allocator should be updated to > 1) Report allocation slack resources when add a new agent. > 2) Allocate allocation slack resources when sending offers. > 3) Update allocation slack resources when update agent, update dynamic > reservations with both framework and http endpoint. -- This message was sent by Atlassian JIRA (v6.3.4#6332)
[jira] [Updated] (MESOS-3931) Do not enable task and executor run on different resources
[ https://issues.apache.org/jira/browse/MESOS-3931?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Guangya Liu updated MESOS-3931: --- Summary: Do not enable task and executor run on different resources (was: Set resource type as ALLOCATION_SLACK for Optimistic Offer) > Do not enable task and executor run on different resources > -- > > Key: MESOS-3931 > URL: https://issues.apache.org/jira/browse/MESOS-3931 > Project: Mesos > Issue Type: Bug >Reporter: Guangya Liu >Assignee: Guangya Liu > > Set resource type as ALLOCATION_SLACK for optimistic Offer so as to > distinguish the revocable resource for allocation slack and usage slack -- This message was sent by Atlassian JIRA (v6.3.4#6332)
[jira] [Updated] (MESOS-3931) Do not enable task and executor run on different resources
[ https://issues.apache.org/jira/browse/MESOS-3931?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Guangya Liu updated MESOS-3931: --- Description: Do not enable task and its executor run on different resources. They must run on same kind of resources, either non-revocable resources, usage slack resources or allocation slack resources. (was: Set resource type as ALLOCATION_SLACK for optimistic Offer so as to distinguish the revocable resource for allocation slack and usage slack) > Do not enable task and executor run on different resources > -- > > Key: MESOS-3931 > URL: https://issues.apache.org/jira/browse/MESOS-3931 > Project: Mesos > Issue Type: Bug >Reporter: Guangya Liu >Assignee: Guangya Liu > > Do not enable task and its executor run on different resources. They must run > on same kind of resources, either non-revocable resources, usage slack > resources or allocation slack resources. -- This message was sent by Atlassian JIRA (v6.3.4#6332)
[jira] [Comment Edited] (MESOS-3931) Do not enable task and executor run on different resources
[ https://issues.apache.org/jira/browse/MESOS-3931?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15051103#comment-15051103 ] Guangya Liu edited comment on MESOS-3931 at 1/3/16 9:26 AM: https://reviews.apache.org/r/41848/ was (Author: gyliu): https://reviews.apache.org/r/40632/ > Do not enable task and executor run on different resources > -- > > Key: MESOS-3931 > URL: https://issues.apache.org/jira/browse/MESOS-3931 > Project: Mesos > Issue Type: Bug >Reporter: Guangya Liu >Assignee: Guangya Liu > > Do not enable task and its executor run on different resources. They must run > on same kind of resources, either non-revocable resources, usage slack > resources or allocation slack resources. -- This message was sent by Atlassian JIRA (v6.3.4#6332)
[jira] [Created] (MESOS-4267) Added helper function to flatten resources.
Guangya Liu created MESOS-4267:
--
Summary: Added helper function to flatten resources.
Key: MESOS-4267
URL: https://issues.apache.org/jira/browse/MESOS-4267
Project: Mesos
Issue Type: Bug
Reporter: Guangya Liu
Assignee: Guangya Liu
We need two helper functions to flatten resources.
{code}
// Returns the reserved resources in fatten by different roles. This is used
// to calculate the initial value for allocation slack.
Resources flattenReserved() const;
// Set resources as SLACK mode, only `ALLOCATION_SLACK` and `USAGE_SLACK`
// are now supported.
Resources flattenSlack(
const Resource::RevocableInfo::Type& type
= Resource::RevocableInfo::ALLOCATION_SLACK) const;
{code}
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
[jira] [Commented] (MESOS-3224) Create a Mesos Contributor Newbie Guide
[
https://issues.apache.org/jira/browse/MESOS-3224?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15080382#comment-15080382
]
haosdent commented on MESOS-3224:
-
{noformat}
{noformat}
is broken, are you forgot to upload image? [~darroyo]
> Create a Mesos Contributor Newbie Guide
> ---
>
> Key: MESOS-3224
> URL: https://issues.apache.org/jira/browse/MESOS-3224
> Project: Mesos
> Issue Type: Documentation
> Components: documentation
>Reporter: Timothy Chen
>Assignee: Diana Arroyo
>
> Currently the website doesn't have a helpful guide for community users to
> know how to start learning to contribute to Mesos, understand the concepts
> and lower the barrier to get involved.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
[jira] [Created] (MESOS-4268) Fix containerizer-internals Posix launcher TBD
haosdent created MESOS-4268: --- Summary: Fix containerizer-internals Posix launcher TBD Key: MESOS-4268 URL: https://issues.apache.org/jira/browse/MESOS-4268 Project: Mesos Issue Type: Bug Reporter: haosdent Assignee: haosdent -- This message was sent by Atlassian JIRA (v6.3.4#6332)
[jira] [Commented] (MESOS-4247) Web UI monitor json loss timestamp
[
https://issues.apache.org/jira/browse/MESOS-4247?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15080400#comment-15080400
]
haosdent commented on MESOS-4247:
-
{noformat}
http http://xxx:5051/monitor/statistics
HTTP/1.1 200 OK
Content-Length: 364
Content-Type: application/json
Date: Sun, 03 Jan 2016 11:21:46 GMT
[
{
"executor_id": "test_sleep",
"executor_name": "Command Executor (Task: test_sleep) (Command: sh -c
'sleep 500')",
"framework_id": "1b63ee6a-4f26-45ca-b314-793fe2136fee-",
"source": "test_sleep",
"statistics": {
"cpus_limit": 1.1,
"cpus_system_time_secs": 0.23,
"cpus_user_time_secs": 0.64,
"mem_limit_bytes": 167772160,
"mem_rss_bytes": 55586816,
"timestamp": 1451820106.77759
}
}
]
{noformat}
I use latest mesos code, the return result have timestamp.
> Web UI monitor json loss timestamp
> --
>
> Key: MESOS-4247
> URL: https://issues.apache.org/jira/browse/MESOS-4247
> Project: Mesos
> Issue Type: Bug
> Components: webui
>Affects Versions: 0.26.0
> Environment: CentOS 7.0
> Mesos 0.26.0
> Marathon 0.13
>Reporter: Michael Zhang
>Priority: Minor
>
> When I use Web UI monitor CPU used for instance, found monitor json callback
> without timestamp and effect cpu used can't be display in web ui
> monitor url refresh:
> http://mesos-server:5051/monitor/statistics?jsonp=angular.callbacks._16w
> 0.22 with timestamp
> {code:xml}
> angular.callbacks._1u([{"executor_id":"x.323375c9-a94c-11e5-8fbb-86eb3605a5cc","executor_name":"Command
> Executor (Task: x.323375c9-a94c-11e5-8fbb-86eb3605a5cc) (Command: sh -c
> 'python
> start...')","framework_id":"20151009-143855-3926863626-5050-24484-","source":"x.323375c9-a94c-11e5-8fbb-86eb3605a5cc","statistics":{"cpus_limit":5.1,"cpus_system_time_secs":28.03,"cpus_user_time_secs":312.57,"mem_limit_bytes":5402263552,"mem_rss_bytes":425009152}},{"executor_id":x-test.e15fa934-a918-11e5-8fbb-86eb3605a5cc","executor_name":"Command
> Executor (Task: x-test.e15fa934-a918-11e5-8fbb-86eb3605a5cc) (Command:
> sh -c 'boinc\/boinc
> ...')","framework_id":"20151009-143855-3926863626-5050-24484-","source":"x-test.e15fa934-a918-11e5-8fbb-86eb3605a5cc","statistics":{"cpus_limit":5.1,"cpus_system_time_secs":133.11,"cpus_user_time_secs":438225.22,"mem_limit_bytes":8422162432,"mem_rss_bytes":38584320}}]);
> {code}
> 0.26 without timestamp
> {code:xml}
> angular.callbacks._46([{"executor_id":"torpedo.781bd286-97fc-11e5-bd8e-0cc47a5235cc","executor_name":"Command
> Executor (Task: xx.781bd286-97fc-11e5-bd8e-0cc47a5235cc) (Command: NO
> EXECUTABLE)","framework_id":"20150727-153101-3456970506-5050-32493-","source":"xx.781bd286-97fc-11e5-bd8e-0cc47a5235cc","statistics":{"cpus_limit":16.1,"cpus_system_time_secs":155083.63,"cpus_user_time_secs":100455.82,"mem_limit_bytes":33587986432,"mem_rss_bytes":667635712,"timestamp":1450946993.81198}},{"executor_id":"x-test.bba5aae5-97fb-11e5-bd8e-0cc47a5235cc","executor_name":"Command
> Executor (Task: x-test.bba5aae5-97fb-11e5-bd8e-0cc47a5235cc) (Command:
> sh -c 'boinc\/boinc
> ...')","framework_id":"20150727-153101-3456970506-5050-32493-","source":"boinc-test.bba5aae5-97fb-11e5-bd8e-0cc47a5235cc","statistics":{"cpus_limit":8.1,"cpus_system_time_secs":1103.55,"cpus_user_time_secs":4865.9,"mem_limit_bytes":8422162432,"mem_rss_bytes":28041216,"timestamp":1450946993.82719}}]);
> {code}
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
[jira] [Updated] (MESOS-4123) Added USAGE_SLACK metrics to snapshot endpoint for master/agent
[
https://issues.apache.org/jira/browse/MESOS-4123?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Guangya Liu updated MESOS-4123:
---
Summary: Added USAGE_SLACK metrics to snapshot endpoint for master/agent
(was: Enable agent/master know resource type is USAGE_SLACK for QoS Controller
related resources)
> Added USAGE_SLACK metrics to snapshot endpoint for master/agent
> ---
>
> Key: MESOS-4123
> URL: https://issues.apache.org/jira/browse/MESOS-4123
> Project: Mesos
> Issue Type: Bug
>Reporter: Guangya Liu
>Assignee: Guangya Liu
>
> Now the master/agent have endpoint to get all revocable resources but the
> current revocable resources are only for QoS controller.
> The current use of those resource are only for some display issues, but we
> may need those APIs in future such as MESOS-2647 , it need to calculate if
> there are enough usage_slack revocable resources before launch a task.
> So I think that we need to update the helper functions of
> {code}_resources_revocable_total{code}
> {code}_resources_revocable_used{code}
> {code}_resources_revocable_percent{code}
> to only get usage_slack revocable resources .
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
[jira] [Updated] (MESOS-4124) Added ALLOCATION_SLACK metrics to snapshot endpoint for master/agent
[ https://issues.apache.org/jira/browse/MESOS-4124?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Guangya Liu updated MESOS-4124: --- Summary: Added ALLOCATION_SLACK metrics to snapshot endpoint for master/agent (was: Enable agent/master know resource type is ALLOCATION_SLACK) > Added ALLOCATION_SLACK metrics to snapshot endpoint for master/agent > > > Key: MESOS-4124 > URL: https://issues.apache.org/jira/browse/MESOS-4124 > Project: Mesos > Issue Type: Bug >Reporter: Guangya Liu >Assignee: Guangya Liu > > Add some helper function in both master and slave to get those allocation > slack resources. Those helper functions can be used by slave checking and > master/slave endpoint checking. -- This message was sent by Atlassian JIRA (v6.3.4#6332)
[jira] [Commented] (MESOS-4060) Memory Leak in Default Mesos UI
[ https://issues.apache.org/jira/browse/MESOS-4060?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15080421#comment-15080421 ] haosdent commented on MESOS-4060: - Hi, [~mandoskippy] Are you use https://addons.mozilla.org/en-US/firefox/addon/tab-data/ Seems I could not reproduce this issue in firefox. > Memory Leak in Default Mesos UI > --- > > Key: MESOS-4060 > URL: https://issues.apache.org/jira/browse/MESOS-4060 > Project: Mesos > Issue Type: Bug > Components: webui >Affects Versions: 0.25.0 > Environment: All >Reporter: John Omernik > > Basically, in using Firefox, I noticed that over time, my Firefox would get > to become unusable when I had Marathon and Mesos WebUIs up and running. At > first, I thought it was a function of my home (Mac) computer and Firefox. But > when I started a PoC for work, and my Windows install of Firefox had the same > issues, I started doing more investigation. > First of all, both at home and at work, my Firefox is only dedicated to > "cluster" related tasks. Thus, I don't have other tabs that are not Cluster > UIs. > My typical setup is to have MapR UI, Mesos UI, Marathon UI, Chronos UI, > Myriad UI, and Yarn UI all up and running. > After about 3-4 hours, my browser would get really slow, and non-responsive. > I'd kill all and start again. Rinse repeat. > So I did some analysis, and basically found a plugin for Firefox that shows > on each tab the amount of memory being used. I found that both Marathon and > Mesos UI were the culprits, and things got really bad after just 2-3 hours. > With those setup, on Windows, I have the following memory usage: > MapR UI: 7.7mb > Myriad UI: 9.8mb > Yarn: 2.7 mb > Chronos 10.2 mb > Marathon: 163 mb > Mesos UI: 463 mb > Both Marathon and Mesos continually climb, slowly, up some, down a few, up > some more, but obviously generally trending up. I guess I wanted to toss it > out here to see if it's something in my settings, or something that others > are seeing. It's a problem for me from a usability standpoint, and I am > guessing that it's one of those things that while not a priority for a > project, should be looked at. -- This message was sent by Atlassian JIRA (v6.3.4#6332)
[jira] [Commented] (MESOS-4060) Memory Leak in Default Mesos UI
[ https://issues.apache.org/jira/browse/MESOS-4060?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15080422#comment-15080422 ] haosdent commented on MESOS-4060: - Maybe you could try use Chrome Javascript Profiles to snapshot memory and show us result. https://developer.chrome.com/devtools/docs/javascript-memory-profiling I also could not reproduce in Chrome. > Memory Leak in Default Mesos UI > --- > > Key: MESOS-4060 > URL: https://issues.apache.org/jira/browse/MESOS-4060 > Project: Mesos > Issue Type: Bug > Components: webui >Affects Versions: 0.25.0 > Environment: All >Reporter: John Omernik > > Basically, in using Firefox, I noticed that over time, my Firefox would get > to become unusable when I had Marathon and Mesos WebUIs up and running. At > first, I thought it was a function of my home (Mac) computer and Firefox. But > when I started a PoC for work, and my Windows install of Firefox had the same > issues, I started doing more investigation. > First of all, both at home and at work, my Firefox is only dedicated to > "cluster" related tasks. Thus, I don't have other tabs that are not Cluster > UIs. > My typical setup is to have MapR UI, Mesos UI, Marathon UI, Chronos UI, > Myriad UI, and Yarn UI all up and running. > After about 3-4 hours, my browser would get really slow, and non-responsive. > I'd kill all and start again. Rinse repeat. > So I did some analysis, and basically found a plugin for Firefox that shows > on each tab the amount of memory being used. I found that both Marathon and > Mesos UI were the culprits, and things got really bad after just 2-3 hours. > With those setup, on Windows, I have the following memory usage: > MapR UI: 7.7mb > Myriad UI: 9.8mb > Yarn: 2.7 mb > Chronos 10.2 mb > Marathon: 163 mb > Mesos UI: 463 mb > Both Marathon and Mesos continually climb, slowly, up some, down a few, up > some more, but obviously generally trending up. I guess I wanted to toss it > out here to see if it's something in my settings, or something that others > are seeing. It's a problem for me from a usability standpoint, and I am > guessing that it's one of those things that while not a priority for a > project, should be looked at. -- This message was sent by Atlassian JIRA (v6.3.4#6332)
[jira] [Commented] (MESOS-3972) Framework CPU counters on slave page are always zero
[ https://issues.apache.org/jira/browse/MESOS-3972?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15080425#comment-15080425 ] haosdent commented on MESOS-3972: - I verify this patch work for me. Thank you very much. > Framework CPU counters on slave page are always zero > > > Key: MESOS-3972 > URL: https://issues.apache.org/jira/browse/MESOS-3972 > Project: Mesos > Issue Type: Bug > Components: webui >Affects Versions: 0.25.0 >Reporter: Ian Babrou > > Don't know why, but system and user are always zero, but total is not. -- This message was sent by Atlassian JIRA (v6.3.4#6332)
[jira] [Updated] (MESOS-4123) Added USAGE_SLACK metrics to snapshot endpoint for master/agent
[
https://issues.apache.org/jira/browse/MESOS-4123?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Guangya Liu updated MESOS-4123:
---
Description:
The master/agent have endpoint to get all revocable resources but the current
revocable resources will include both usage slack and allocation slack
resources.
It is better to put usage slack metrics to
was:
Now the master/agent have endpoint to get all revocable resources but the
current revocable resources are only for QoS controller.
The current use of those resource are only for some display issues, but we may
need those APIs in future such as MESOS-2647 , it need to calculate if there
are enough usage_slack revocable resources before launch a task.
So I think that we need to update the helper functions of
{code}_resources_revocable_total{code}
{code}_resources_revocable_used{code}
{code}_resources_revocable_percent{code}
to only get usage_slack revocable resources .
> Added USAGE_SLACK metrics to snapshot endpoint for master/agent
> ---
>
> Key: MESOS-4123
> URL: https://issues.apache.org/jira/browse/MESOS-4123
> Project: Mesos
> Issue Type: Bug
>Reporter: Guangya Liu
>Assignee: Guangya Liu
>
> The master/agent have endpoint to get all revocable resources but the current
> revocable resources will include both usage slack and allocation slack
> resources.
> It is better to put usage slack metrics to
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
[jira] [Updated] (MESOS-4123) Added USAGE_SLACK metrics to snapshot endpoint for master/agent
[ https://issues.apache.org/jira/browse/MESOS-4123?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Guangya Liu updated MESOS-4123: --- Description: The master/agent have endpoint to get all revocable resources but the current revocable resources will include both usage slack and allocation slack resources. It is better to put usage slack metrics to snapshot endpoint for master/agent. was: The master/agent have endpoint to get all revocable resources but the current revocable resources will include both usage slack and allocation slack resources. It is better to put usage slack metrics to > Added USAGE_SLACK metrics to snapshot endpoint for master/agent > --- > > Key: MESOS-4123 > URL: https://issues.apache.org/jira/browse/MESOS-4123 > Project: Mesos > Issue Type: Bug >Reporter: Guangya Liu >Assignee: Guangya Liu > > The master/agent have endpoint to get all revocable resources but the current > revocable resources will include both usage slack and allocation slack > resources. > It is better to put usage slack metrics to snapshot endpoint for master/agent. -- This message was sent by Atlassian JIRA (v6.3.4#6332)
[jira] [Updated] (MESOS-4124) Added ALLOCATION_SLACK metrics to snapshot endpoint for master/agent
[ https://issues.apache.org/jira/browse/MESOS-4124?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Guangya Liu updated MESOS-4124: --- Description: The master/agent have endpoint to get all revocable resources but the current revocable resources will include both usage slack and allocation slack resources. It is better to put allocation slack metrics to snapshot endpoint for master/agent. was:Add some helper function in both master and slave to get those allocation slack resources. Those helper functions can be used by slave checking and master/slave endpoint checking. > Added ALLOCATION_SLACK metrics to snapshot endpoint for master/agent > > > Key: MESOS-4124 > URL: https://issues.apache.org/jira/browse/MESOS-4124 > Project: Mesos > Issue Type: Bug >Reporter: Guangya Liu >Assignee: Guangya Liu > > The master/agent have endpoint to get all revocable resources but the current > revocable resources will include both usage slack and allocation slack > resources. > It is better to put allocation slack metrics to snapshot endpoint for > master/agent. -- This message was sent by Atlassian JIRA (v6.3.4#6332)
[jira] [Comment Edited] (MESOS-4124) Added ALLOCATION_SLACK metrics to snapshot endpoint for master/agent
[ https://issues.apache.org/jira/browse/MESOS-4124?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15052461#comment-15052461 ] Guangya Liu edited comment on MESOS-4124 at 1/3/16 1:40 PM: https://reviews.apache.org/r/41250/ https://reviews.apache.org/r/41254 was (Author: gyliu): https://reviews.apache.org/r/41250/ https://reviews.apache.org/r/40529/ > Added ALLOCATION_SLACK metrics to snapshot endpoint for master/agent > > > Key: MESOS-4124 > URL: https://issues.apache.org/jira/browse/MESOS-4124 > Project: Mesos > Issue Type: Bug >Reporter: Guangya Liu >Assignee: Guangya Liu > > The master/agent have endpoint to get all revocable resources but the current > revocable resources will include both usage slack and allocation slack > resources. > It is better to put allocation slack metrics to snapshot endpoint for > master/agent. -- This message was sent by Atlassian JIRA (v6.3.4#6332)
[jira] [Commented] (MESOS-3971) Running task counters in mesos UI is racy.
[ https://issues.apache.org/jira/browse/MESOS-3971?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15080434#comment-15080434 ] haosdent commented on MESOS-3971: - Hi, [~bobrik] Could you help provide more details? I could see $scope.staged_tasks used for master while $scope.state.staged_tasks used for slave. It should not be racy. > Running task counters in mesos UI is racy. > -- > > Key: MESOS-3971 > URL: https://issues.apache.org/jira/browse/MESOS-3971 > Project: Mesos > Issue Type: Bug > Components: webui >Affects Versions: 0.25.0 >Reporter: Ian Babrou > > On slave's page task counters in the left panel are racy. > The reason for that is: in src/webui/master/static/js/controllers.js > properties like "staged_tasks" are populated from both master and slave > states. -- This message was sent by Atlassian JIRA (v6.3.4#6332)
[jira] [Created] (MESOS-4269) Minor typo in src/linux/cgroups.cpp
Ryuichi Okumura created MESOS-4269: -- Summary: Minor typo in src/linux/cgroups.cpp Key: MESOS-4269 URL: https://issues.apache.org/jira/browse/MESOS-4269 Project: Mesos Issue Type: Bug Reporter: Ryuichi Okumura Assignee: Ryuichi Okumura Priority: Minor There is a typo in the "src/linux/cgroups.cpp" as follows. https://github.com/apache/mesos/blob/765c025dd43e04360b29c19bd9a66837954c5a20/src/linux/cgroups.cpp#L1438 -- This message was sent by Atlassian JIRA (v6.3.4#6332)
[jira] [Commented] (MESOS-4269) Minor typo in src/linux/cgroups.cpp
[ https://issues.apache.org/jira/browse/MESOS-4269?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15080439#comment-15080439 ] Ryuichi Okumura commented on MESOS-4269: Ready for review. https://reviews.apache.org/r/41863/ > Minor typo in src/linux/cgroups.cpp > --- > > Key: MESOS-4269 > URL: https://issues.apache.org/jira/browse/MESOS-4269 > Project: Mesos > Issue Type: Bug >Reporter: Ryuichi Okumura >Assignee: Ryuichi Okumura >Priority: Minor > > There is a typo in the "src/linux/cgroups.cpp" as follows. > https://github.com/apache/mesos/blob/765c025dd43e04360b29c19bd9a66837954c5a20/src/linux/cgroups.cpp#L1438 -- This message was sent by Atlassian JIRA (v6.3.4#6332)
[jira] [Commented] (MESOS-3702) Mesos UI -> Slaves shows not all resources when roles applied
[
https://issues.apache.org/jira/browse/MESOS-3702?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15080440#comment-15080440
]
haosdent commented on MESOS-3702:
-
Hi, [~tpolekhin] webui not care resources role so far. Are you sure your slave
have more than 4 cpu? Mesos webui just add up all slaves resources from state
endpoint.
{noformat}
~ http http://xxx:5050/state
...
slaves: [{resources: ...}]
{noformat}
> Mesos UI -> Slaves shows not all resources when roles applied
> --
>
> Key: MESOS-3702
> URL: https://issues.apache.org/jira/browse/MESOS-3702
> Project: Mesos
> Issue Type: Bug
> Components: webui
>Affects Versions: 0.22.1
>Reporter: Tymofii
>Priority: Trivial
> Labels: webui
>
> When using Mesos Roles for splitting resources Mesos Web UI not showing them
> right.
> I have 6 roles configured on master node:
> MESOS_ROLES=slave_public,cassandra,elastic,hdfs,kafka,generic
> The mesos slave resource configuration:
> MESOS_RESOURCES="cpus(cassandra):2;mem(cassandra):16384;disk(cassandra):1048576;ports(cassandra):[9000-];cpus(elastic):4;mem(elastic):16384;disk(elastic):2097152;ports(elastic):[5052-5053];cpus(hdfs):4;mem(hdfs):16384;disk(hdfs):4194304;ports(hdfs):[8000-8999,5-50999];cpus(kafka):2;mem(kafka):16384;disk(kafka):10;ports(kafka):[1025-1026];cpus(generic):4;mem(generic):65536;disk(generic):3145718;ports(generic):[1027-2180,2182-3887,3889-5049,5054-7999,1-4,51000-65535];"
> When i open up Mesos UI -> Slaves, it shows only 4 cpu, 64 ram and 3TB disk
> for each slave, which is not all resources, just 'generic' one
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
[jira] [Commented] (MESOS-2635) Web UI Display Bug when starting lots of tasks with small cpu value
[ https://issues.apache.org/jira/browse/MESOS-2635?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15080441#comment-15080441 ] haosdent commented on MESOS-2635: - The patch update framework tab. https://reviews.apache.org/r/41864/ > Web UI Display Bug when starting lots of tasks with small cpu value > --- > > Key: MESOS-2635 > URL: https://issues.apache.org/jira/browse/MESOS-2635 > Project: Mesos > Issue Type: Bug > Components: webui >Affects Versions: 0.22.0 >Reporter: Lukas Loesche >Assignee: haosdent >Priority: Minor > Fix For: 0.25.0 > > Attachments: Screen Shot 2015-04-17 at 18.53.54.png, fixed_after.png, > fixed_before.png > > > During Marathon scale testing we encountered a display bug where the Mesos > Web UI would show -4.064304448547773e-11 Idle CPUs in the Resources overview. > To reproduce, we started: > "cmd": "python -m SimpleHTTPServer $PORT0", > "cpus": 0.0001, > "instances": 2000, > "mem": 32, -- This message was sent by Atlassian JIRA (v6.3.4#6332)
[jira] [Commented] (MESOS-2895) add static html serving to web ui
[ https://issues.apache.org/jira/browse/MESOS-2895?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15080442#comment-15080442 ] haosdent commented on MESOS-2895: - Hi, the webui_dir in Mesos master is used for serve static files, is this work for solving your problem? > add static html serving to web ui > - > > Key: MESOS-2895 > URL: https://issues.apache.org/jira/browse/MESOS-2895 > Project: Mesos > Issue Type: Improvement > Components: webui >Reporter: brian wickman > > We have a use-case in Aurora where we'd like the executor to generate static > html pages that can be served directly by the master, saving us from having > to allocate a port and run our own webserver from the task. Right now we > only have the "pailer" which allows us to tail files and download links. > We'd like for an endpoint that serves files with proper mime types. -- This message was sent by Atlassian JIRA (v6.3.4#6332)
[jira] [Updated] (MESOS-2585) Use full width for mesos div.container
[ https://issues.apache.org/jira/browse/MESOS-2585?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] haosdent updated MESOS-2585: Attachment: github_full_width.png > Use full width for mesos div.container > -- > > Key: MESOS-2585 > URL: https://issues.apache.org/jira/browse/MESOS-2585 > Project: Mesos > Issue Type: Improvement > Components: webui >Reporter: Alson Kemp >Priority: Trivial > Attachments: After (patch 2).png, Narrow (current).png, Wide > (patched).png, github_full_width.png > > > I've patched our Mesos installation so that the webui takes up the full page > width and is much nicer to look at on large monitors. It's a small change. > If y'all want me to submit a PR with the update, I'll do so. > Before: > !https://issues.apache.org/jira/secure/attachment/12708818/Narrow%20%28current%29.png|width=800! > After: > !https://issues.apache.org/jira/secure/attachment/12708861/After%20%28patch%202%29.png|width=800! -- This message was sent by Atlassian JIRA (v6.3.4#6332)
[jira] [Commented] (MESOS-2585) Use full width for mesos div.container
[
https://issues.apache.org/jira/browse/MESOS-2585?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15080449#comment-15080449
]
haosdent commented on MESOS-2585:
-
I think we could follow what github do here
{noformat}
body.split-diff .container {
width: 100%;
padding-left: 30px;
padding-right: 30px
}
{noformat}
!https://issues.apache.org/jira/secure/attachment/12780235/github_full_width.png!
> Use full width for mesos div.container
> --
>
> Key: MESOS-2585
> URL: https://issues.apache.org/jira/browse/MESOS-2585
> Project: Mesos
> Issue Type: Improvement
> Components: webui
>Reporter: Alson Kemp
>Priority: Trivial
> Attachments: After (patch 2).png, Narrow (current).png, Wide
> (patched).png, github_full_width.png
>
>
> I've patched our Mesos installation so that the webui takes up the full page
> width and is much nicer to look at on large monitors. It's a small change.
> If y'all want me to submit a PR with the update, I'll do so.
> Before:
> !https://issues.apache.org/jira/secure/attachment/12708818/Narrow%20%28current%29.png|width=800!
> After:
> !https://issues.apache.org/jira/secure/attachment/12708861/After%20%28patch%202%29.png|width=800!
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
[jira] [Assigned] (MESOS-2585) Use full width for mesos div.container
[ https://issues.apache.org/jira/browse/MESOS-2585?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] haosdent reassigned MESOS-2585: --- Assignee: haosdent > Use full width for mesos div.container > -- > > Key: MESOS-2585 > URL: https://issues.apache.org/jira/browse/MESOS-2585 > Project: Mesos > Issue Type: Improvement > Components: webui >Reporter: Alson Kemp >Assignee: haosdent >Priority: Trivial > Attachments: After (patch 2).png, Narrow (current).png, Wide > (patched).png, github_full_width.png > > > I've patched our Mesos installation so that the webui takes up the full page > width and is much nicer to look at on large monitors. It's a small change. > If y'all want me to submit a PR with the update, I'll do so. > Before: > !https://issues.apache.org/jira/secure/attachment/12708818/Narrow%20%28current%29.png|width=800! > After: > !https://issues.apache.org/jira/secure/attachment/12708861/After%20%28patch%202%29.png|width=800! -- This message was sent by Atlassian JIRA (v6.3.4#6332)
[jira] [Commented] (MESOS-2446) Show 'checkpoint' field in framework view in the Mesos Web UI
[ https://issues.apache.org/jira/browse/MESOS-2446?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15080451#comment-15080451 ] haosdent commented on MESOS-2446: - Hi, [~nnielsen] Should show them in ./src/webui/master/static/framework.html or ./src/webui/master/static/frameworks.html? Or both? Seems add a new column to frameworks table looks not nice. > Show 'checkpoint' field in framework view in the Mesos Web UI > - > > Key: MESOS-2446 > URL: https://issues.apache.org/jira/browse/MESOS-2446 > Project: Mesos > Issue Type: Improvement > Components: webui >Reporter: Niklas Quarfot Nielsen >Priority: Trivial > -- This message was sent by Atlassian JIRA (v6.3.4#6332)
[jira] [Commented] (MESOS-1976) Sandbox browse UI has path which is not selectable
[
https://issues.apache.org/jira/browse/MESOS-1976?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15080488#comment-15080488
]
haosdent commented on MESOS-1976:
-
Hi, this is a expected behaviour when use bootstrap breadcrumbs
http://getbootstrap.com/components/#breadcrumbs
> Sandbox browse UI has path which is not selectable
> --
>
> Key: MESOS-1976
> URL: https://issues.apache.org/jira/browse/MESOS-1976
> Project: Mesos
> Issue Type: Bug
> Components: webui
>Affects Versions: 0.20.1
>Reporter: Steven Schlansker
>Priority: Minor
>
> The Sandbox UI displays the path being browsed as a series of links. It is
> not possible to copy the path from this, it ends up being formatted as e.g.
> {code}
> mnt
> mesos
> slaves
> 20141022-230146-2500085258-5050-1554-3
> frameworks
> Singularity
> executors
> ci-discovery-singularity-bridge-steven.2014.10.21T21.00.04-1414092693380-2-10-us_west_2a
> runs
> 554eebb3-126d-42bd-95c2-aa8282b05522
> {code}
> instead of the expected
> {code}
> /mnt/mesos/slaves/20141022-230146-2500085258-5050-1554-3/frameworks/Singularity/executors/ci-discovery-singularity-bridge-steven.2014.10.21T21.00.04-1414092693380-2-10-us_west_2a/runs/554eebb3-126d-42bd-95c2-aa8282b05522
>
> {code}
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
[jira] [Commented] (MESOS-3971) Running task counters in mesos UI is racy.
[ https://issues.apache.org/jira/browse/MESOS-3971?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15080505#comment-15080505 ] Ian Babrou commented on MESOS-3971: --- To validate raciness just open slave's page, look at "Slave: " line and refresh a few times. Eventually you'll see that the value comes from master first, then it's replaced by slave's hostname. http://imgur.com/BDzzKM2 — no tasks counters shown, could be a related issue. Refreshing fixes this eventually. I made a video where both issues are visible: https://drive.google.com/file/d/0B0ykRvtbqOg3VzRrODI3QkV2TTA/view?usp=sharing > Running task counters in mesos UI is racy. > -- > > Key: MESOS-3971 > URL: https://issues.apache.org/jira/browse/MESOS-3971 > Project: Mesos > Issue Type: Bug > Components: webui >Affects Versions: 0.25.0 >Reporter: Ian Babrou > > On slave's page task counters in the left panel are racy. > The reason for that is: in src/webui/master/static/js/controllers.js > properties like "staged_tasks" are populated from both master and slave > states. -- This message was sent by Atlassian JIRA (v6.3.4#6332)
[jira] [Commented] (MESOS-3809) Expose advertise_ip and advertise_port as command line options in mesos slave
[ https://issues.apache.org/jira/browse/MESOS-3809?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15080520#comment-15080520 ] Anindya Sinha commented on MESOS-3809: -- Thanks [~metskem]. Looks like that this commit did not actually make it into 0.26. It does show up in the file "CHANGELOG" in the tag "0.26.0" (as well as the tag "0.26.0-rc2" when it was first introduced), but the actual commit did not make it. I will follow up and update accordingly. For the tag 0.26: https://github.com/apache/mesos/blob/0.26.0/CHANGELOG#L58: This adds MESOS-3809 into this release (as in CHANGELOG). But the change in docs/configuration.md (https://github.com/apache/mesos/blob/0.26.0/docs/configuration.md) and src/slave/main.cpp (https://github.com/apache/mesos/blob/0.26.0/src/slave/main.cpp) is missing based on original commit in master which is: bb5c44a7a6fead84789c933a9aa3e1a8cf35722d Added 2 slave flags --advertise_ip and --advertise_port. > Expose advertise_ip and advertise_port as command line options in mesos slave > - > > Key: MESOS-3809 > URL: https://issues.apache.org/jira/browse/MESOS-3809 > Project: Mesos > Issue Type: Bug > Components: slave >Affects Versions: 0.25.0 >Reporter: Anindya Sinha >Assignee: Anindya Sinha >Priority: Minor > Labels: mesosphere > Fix For: 0.26.0 > > > advertise_ip and advertise_port are exposed as mesos master command line args > (MESOS-809). But the following use case makes it a candidate for adding as > command line args in mesos slave as well. > On Tue, Oct 27, 2015 at 7:43 PM, Xiaodong Zhang wrote: > It works! Thanks a lot. > 发件人: haosdent > 答复: "u...@mesos.apache.org" > 日期: 2015年10月28日 星期三 上午10:23 > 至: user > 主题: Re: How to tell master which ip to connect. > Do you try `export LIBPROCESS_ADVERTISE_IP=xxx` and > `LIBPROCESS_ADVERTISE_PORT` when start slave? > On Wed, Oct 28, 2015 at 10:16 AM, Xiaodong Zhang wrote: > Hi teams: > My scenarios is like this: > My master nodes were deployed in AWS. My slaves were in AZURE.So they > communicate via public ip. > I got trouble when slaves try to register to master. > Now slaves can get master’s public ip address,and can send register > request.But they can only send there private ip to master.(Because they don’t > know there public ip,thus they can’t not bind a public ip via —ip flag), > thus masters can’t connect slaves.How can the slave to tell master which ip > master should connect(I can’t find any flags like —advertise_ip in master). -- This message was sent by Atlassian JIRA (v6.3.4#6332)
[jira] [Updated] (MESOS-3897) Identify and implement test cases for verifying eviction logic in the agent
[ https://issues.apache.org/jira/browse/MESOS-3897?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Klaus Ma updated MESOS-3897: Summary: Identify and implement test cases for verifying eviction logic in the agent (was: Identify and implement test cases for verifying eviction logic in the allocator.) > Identify and implement test cases for verifying eviction logic in the agent > --- > > Key: MESOS-3897 > URL: https://issues.apache.org/jira/browse/MESOS-3897 > Project: Mesos > Issue Type: Bug >Reporter: Artem Harutyunyan >Assignee: Artem Harutyunyan > Labels: mesosphere > -- This message was sent by Atlassian JIRA (v6.3.4#6332)
[jira] [Assigned] (MESOS-3897) Identify and implement test cases for verifying eviction logic in the agent
[ https://issues.apache.org/jira/browse/MESOS-3897?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Klaus Ma reassigned MESOS-3897: --- Assignee: Klaus Ma (was: Artem Harutyunyan) > Identify and implement test cases for verifying eviction logic in the agent > --- > > Key: MESOS-3897 > URL: https://issues.apache.org/jira/browse/MESOS-3897 > Project: Mesos > Issue Type: Bug >Reporter: Artem Harutyunyan >Assignee: Klaus Ma > Labels: mesosphere > -- This message was sent by Atlassian JIRA (v6.3.4#6332)
[jira] [Assigned] (MESOS-3898) Identify and implement test cases for handling a race between optimistic lender and tenant offers.
[ https://issues.apache.org/jira/browse/MESOS-3898?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Klaus Ma reassigned MESOS-3898: --- Assignee: Klaus Ma (was: Artem Harutyunyan) > Identify and implement test cases for handling a race between optimistic > lender and tenant offers. > -- > > Key: MESOS-3898 > URL: https://issues.apache.org/jira/browse/MESOS-3898 > Project: Mesos > Issue Type: Bug >Reporter: Artem Harutyunyan >Assignee: Klaus Ma > Labels: mesosphere > > An example is the when lender launches the task on an agent followed by a > borrower launching a task on the same agent before the optimistic offer is > rescinded. -- This message was sent by Atlassian JIRA (v6.3.4#6332)
[jira] [Assigned] (MESOS-3893) Implement tests for verifying allocator resource math.
[ https://issues.apache.org/jira/browse/MESOS-3893?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Guangya Liu reassigned MESOS-3893: -- Assignee: Guangya Liu (was: Artem Harutyunyan) > Implement tests for verifying allocator resource math. > -- > > Key: MESOS-3893 > URL: https://issues.apache.org/jira/browse/MESOS-3893 > Project: Mesos > Issue Type: Bug >Reporter: Artem Harutyunyan >Assignee: Guangya Liu > Labels: mesosphere > > Write a test to ensure that the allocator performs the reservation slack > calculations correctly. -- This message was sent by Atlassian JIRA (v6.3.4#6332)
[jira] [Commented] (MESOS-3896) Add accounting for reservation slack in the allocator.
[
https://issues.apache.org/jira/browse/MESOS-3896?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15080661#comment-15080661
]
Guangya Liu commented on MESOS-3896:
I think that this one was already covered in
https://issues.apache.org/jira/browse/MESOS-4145
> Add accounting for reservation slack in the allocator.
> --
>
> Key: MESOS-3896
> URL: https://issues.apache.org/jira/browse/MESOS-3896
> Project: Mesos
> Issue Type: Bug
>Reporter: Artem Harutyunyan
>Assignee: Klaus Ma
> Labels: mesosphere
>
> MESOS-XXX: Optimsistic accounter
> {code}
> class HierarchicalAllocatorProcess
> {
> struct Slave
> {
> ...
> struct Optimistic
> {
> Resources total; // The total allocation slack resources
> Resources allocated; // The allocated allocation slack resources
> };
>
> Optimistic optimistic;
> };
> }
> {code}
> MESOS-4146: flatten & allocationSlack for Optimistic Offer
> {code}
> class Resources
> {
> // Returns a Resources object with the same amount of each resource
> // type as these Resources, but with all Resource objects marked as
> // the specified `RevocableInfo::Type`; the other attribute is not
> // affected.
> Resources flatten(Resource::RevocableInfo::Type type);
> // Return a Resources object that:
> // - if role is given, the resources did not include role's reserved
> // resources.
> // - the resources's revocable type is `ALLOCATION_SLACK`
> // - the role of resources is set to "*"
> Resources allocationSlack(Option role = None());
> }
> {code}
> MESOS-XXX: Allocate the allocation_slack resources to framework
> {code}
> void HierarchicalAllocatorProcess::allocate(
> const hashset& slaveIds_)
> {
> foreach slave; foreach role; foreach framework
> {
> Resource optimistic;
> if (framework.revocable) {
> Resources total =
> slaves[slaveId].optimistic.total.allocationSlack(role);
> optimistic = total - slaves[slaveId].optimistic.allocated;
> }
> ...
> offerable[frameworkId][slaveId] += resources + optimistic;
> ...
> slaves[slaveId].optimistic.allocated += optimistic;
> }
> }
> {code}
>
> Here's some consideration about `ALLOCATION_SLACK`:
> 1. 'Old' resources (available/total) did not include ALLOCATION_SLACK
> 2. After `Quota`, `remainingClusterResources.contains` should not check
> ALLOCATION_SLACK; if there no enough resources, master can still offer
> ALLOCATION_SALCK resources.
> 3. In sorter, it'll not include ALLOCATION_SLACK; as those resources are
> borrowed from other role/framework
> 4. If either normal resources or ALLOCATION_SLACK resources are
> allocable/!filtered, it can be offered to framework
> 5. Currently, allocator will assign all ALLOCATION_SALCK resources in slave
> to one framework
> MESOS-XXX: Update ALLOCATION_SLACK for dynamic reservation (updateAllocation)
> {code}
> void HierarchicalAllocatorProcess::updateAllocation(
> const FrameworkID& frameworkId,
> const SlaveID& slaveId,
> const vector& operations)
> {
> ...
> Try updatedOptimistic =
> slaves[slaveId].optimistic.total.apply(operations);
> CHECK_SOME(updatedTotal);
> slaves[slaveId].optimistic.total =
>
> updatedOptimistic.get().stateless().reserved().flatten(ALLOCATION_SLACK);
> ...
> }
> {code}
>
> MESOS-XXX: Add ALLOCATION_SLACK when slaver register/re-register (addSlave)
> {code}
> void HierarchicalAllocatorProcess::addSlave(
> const SlaveID& slaveId,
> const SlaveInfo& slaveInfo,
> const Option& unavailability,
> const Resources& total,
> const hashmap& used)
> {
> ...
> slaves[slaveId].optimistic.total =
> total.stateless().reserved().flatten(ALLOCATION_SLACK);
> ...
> }
> {code}
>
> No need to handle `removeSlave`, it'll all related info from `slaves`
> including `optimistic`.
> MESOS-XXX: return resources to allocator (recoverResources)
> {code}
> void HierarchicalAllocatorProcess::recoverResources(
> const FrameworkID& frameworkId,
> const SlaveID& slaveId,
> const Resources& resources,
> const Option& filters)
> {
> if (slaves.contains(slaveId))
> {
> ...
> slaves[slaveId].optimistic.allocated -= resources.allocationSlack();
> ...
> }
> }
> {code}
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
[jira] [Updated] (MESOS-3972) Framework CPU counters on slave page are always zero
[ https://issues.apache.org/jira/browse/MESOS-3972?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Till Toenshoff updated MESOS-3972: -- Shepherd: Till Toenshoff > Framework CPU counters on slave page are always zero > > > Key: MESOS-3972 > URL: https://issues.apache.org/jira/browse/MESOS-3972 > Project: Mesos > Issue Type: Bug > Components: webui >Affects Versions: 0.25.0 >Reporter: Ian Babrou > > Don't know why, but system and user are always zero, but total is not. -- This message was sent by Atlassian JIRA (v6.3.4#6332)
[jira] [Updated] (MESOS-3972) Framework CPU counters on slave page are always zero
[ https://issues.apache.org/jira/browse/MESOS-3972?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Till Toenshoff updated MESOS-3972: -- Assignee: Ian Babrou > Framework CPU counters on slave page are always zero > > > Key: MESOS-3972 > URL: https://issues.apache.org/jira/browse/MESOS-3972 > Project: Mesos > Issue Type: Bug > Components: webui >Affects Versions: 0.25.0 >Reporter: Ian Babrou >Assignee: Ian Babrou > > Don't know why, but system and user are always zero, but total is not. -- This message was sent by Atlassian JIRA (v6.3.4#6332)
[jira] [Commented] (MESOS-3763) Need for http::put request method
[ https://issues.apache.org/jira/browse/MESOS-3763?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15080721#comment-15080721 ] Yongqiao Wang commented on MESOS-3763: -- [~js84] and [~bmahler], I think we already have a function process::http::request(const Request& request, bool streamedResponse) to meet our requirement. Can we use it directly? > Need for http::put request method > - > > Key: MESOS-3763 > URL: https://issues.apache.org/jira/browse/MESOS-3763 > Project: Mesos > Issue Type: Task >Reporter: Joerg Schad >Assignee: Joerg Schad >Priority: Minor > Labels: mesosphere > > As we decided to create a more restful api for managing Quota request. > Therefore we also want to use the HTTP put request and hence need to enable > the libprocess/http to send put request besides get and post requests. -- This message was sent by Atlassian JIRA (v6.3.4#6332)
