Robert Casey
Hi MUSTANG users-
In addition to the aforementioned updates, we are to de-list the timing_drift metric from MUSTANG during the update process, as the IRIS Quality Assurance Advisory Committee agreed to the case made that it is currently ineffective in properly and consistently detecting timing drift at the indicated station. The QAAC will continue to revisit this metric to seek alternative algorithms that can better serve to make an appropriate measurement in this regard.
As always, we welcome feedback on our metrics and the algorithms used, and will continue to develop, enhance, and critique our metrics code to ensure they are functioning in the most effective way possible. In addition to this list, we also have a general discussion list for MUSTANG that you can sign up to by going to our Message Center: http://ds.iris.edu/message-center/topic/mustang-users/
-Rob
On Oct 24, 2016, at 4:24 PM, Robert Casey <rob<at>iris.washington.edu> wrote:
Hi MUSTANG users-
Feature fixes and tests have been conducted for an important update to MUSTANG, specifically the web services that you use when getting access to the tabular (non-PSD) metrics. We are ready to put this into production tomorrow morning, October 25th, at 10 AM PDT. Problems are not anticipated with the switchover, and it may be entirely transparent to users, but I want to make sure everyone has a heads up just in case there is some disruption in service.
What you can expect in this new version should be to your liking. A summary of what is offered is the following:
Release version 1.2.0 (2016-10-25)
reworked code base to improve efficiencies
support for glob-style wildcards in all SNCL query terms
support for virtual network queries
support for ‘--’ location code queries
support for comma-separated query terms
support for ‘start’ and ‘end’ date-time terms
support for nodata term to send user specified HTTP code when no metrics are found
Improvements to helpful error message framework
LDDATE field is truncated to seconds instead of milli/microseconds
Allow partial matches for channel and location pairs found in certain metrics
Some metric values truncated to a specific precision
The on-line documentation for these features has been drafted for the 'measurements' service. The 'targets' and 'metrics' services will have their documents updated as well. Issues with the URL builder interface will be addressed as needed.
Thank you for your patience as we get this new install into production.
-Rob
-
Sorry for the disruption in service. Discovered an embedded config setting that shouldn't have been there. Redeploying a fix so it can connect to the production database.
-Rob
On Oct 25, 2016, at 9:29 AM, Robert Casey <rob<at>iris.washington.edu> wrote:
Hi MUSTANG users-
In addition to the aforementioned updates, we are to de-list the timing_drift metric from MUSTANG during the update process, as the IRIS Quality Assurance Advisory Committee agreed to the case made that it is currently ineffective in properly and consistently detecting timing drift at the indicated station. The QAAC will continue to revisit this metric to seek alternative algorithms that can better serve to make an appropriate measurement in this regard.
As always, we welcome feedback on our metrics and the algorithms used, and will continue to develop, enhance, and critique our metrics code to ensure they are functioning in the most effective way possible. In addition to this list, we also have a general discussion list for MUSTANG that you can sign up to by going to our Message Center: http://ds.iris.edu/message-center/topic/mustang-users/
-Rob
On Oct 24, 2016, at 4:24 PM, Robert Casey <rob<at>iris.washington.edu <rob<at>iris.washington.edu>> wrote:
Hi MUSTANG users-
Feature fixes and tests have been conducted for an important update to MUSTANG, specifically the web services that you use when getting access to the tabular (non-PSD) metrics. We are ready to put this into production tomorrow morning, October 25th, at 10 AM PDT. Problems are not anticipated with the switchover, and it may be entirely transparent to users, but I want to make sure everyone has a heads up just in case there is some disruption in service.
What you can expect in this new version should be to your liking. A summary of what is offered is the following:
Release version 1.2.0 (2016-10-25)
reworked code base to improve efficiencies
support for glob-style wildcards in all SNCL query terms
support for virtual network queries
support for ‘--’ location code queries
support for comma-separated query terms
support for ‘start’ and ‘end’ date-time terms
support for nodata term to send user specified HTTP code when no metrics are found
Improvements to helpful error message framework
LDDATE field is truncated to seconds instead of milli/microseconds
Allow partial matches for channel and location pairs found in certain metrics
Some metric values truncated to a specific precision
The on-line documentation for these features has been drafted for the 'measurements' service. The 'targets' and 'metrics' services will have their documents updated as well. Issues with the URL builder interface will be addressed as needed.
Thank you for your patience as we get this new install into production.
-Rob
-
MUSTANG Service has been restored. Documentation for these web services will get installed over the course of the day. All functionality should be backward compatible to the previous version of service.
Thank you for your patience.
-Rob
On Oct 25, 2016, at 10:24 AM, Robert Casey <rob<at>iris.washington.edu> wrote:
Sorry for the disruption in service. Discovered an embedded config setting that shouldn't have been there. Redeploying a fix so it can connect to the production database.
-Rob
On Oct 25, 2016, at 9:29 AM, Robert Casey <rob<at>iris.washington.edu <rob<at>iris.washington.edu>> wrote:
Hi MUSTANG users-
In addition to the aforementioned updates, we are to de-list the timing_drift metric from MUSTANG during the update process, as the IRIS Quality Assurance Advisory Committee agreed to the case made that it is currently ineffective in properly and consistently detecting timing drift at the indicated station. The QAAC will continue to revisit this metric to seek alternative algorithms that can better serve to make an appropriate measurement in this regard.
As always, we welcome feedback on our metrics and the algorithms used, and will continue to develop, enhance, and critique our metrics code to ensure they are functioning in the most effective way possible. In addition to this list, we also have a general discussion list for MUSTANG that you can sign up to by going to our Message Center: http://ds.iris.edu/message-center/topic/mustang-users/
-Rob
On Oct 24, 2016, at 4:24 PM, Robert Casey <rob<at>iris.washington.edu <rob<at>iris.washington.edu>> wrote:
Hi MUSTANG users-
Feature fixes and tests have been conducted for an important update to MUSTANG, specifically the web services that you use when getting access to the tabular (non-PSD) metrics. We are ready to put this into production tomorrow morning, October 25th, at 10 AM PDT. Problems are not anticipated with the switchover, and it may be entirely transparent to users, but I want to make sure everyone has a heads up just in case there is some disruption in service.
What you can expect in this new version should be to your liking. A summary of what is offered is the following:
Release version 1.2.0 (2016-10-25)
reworked code base to improve efficiencies
support for glob-style wildcards in all SNCL query terms
support for virtual network queries
support for ‘--’ location code queries
support for comma-separated query terms
support for ‘start’ and ‘end’ date-time terms
support for nodata term to send user specified HTTP code when no metrics are found
Improvements to helpful error message framework
LDDATE field is truncated to seconds instead of milli/microseconds
Allow partial matches for channel and location pairs found in certain metrics
Some metric values truncated to a specific precision
The on-line documentation for these features has been drafted for the 'measurements' service. The 'targets' and 'metrics' services will have their documents updated as well. Issues with the URL builder interface will be addressed as needed.
Thank you for your patience as we get this new install into production.
-Rob
-
MUSTANG Users-
The documentation for http://service.iris.edu/mustang/measurements/1 is up for your review. Please take note of the new usage features and try them out. 'metrics' and 'targets' web services will get an update soon.
-Rob
On Oct 25, 2016, at 10:52 AM, Robert Casey <rob<at>iris.washington.edu> wrote:
MUSTANG Service has been restored. Documentation for these web services will get installed over the course of the day. All functionality should be backward compatible to the previous version of service.
Thank you for your patience.
-Rob
On Oct 25, 2016, at 10:24 AM, Robert Casey <rob<at>iris.washington.edu <rob<at>iris.washington.edu>> wrote:
Sorry for the disruption in service. Discovered an embedded config setting that shouldn't have been there. Redeploying a fix so it can connect to the production database.
-Rob
On Oct 25, 2016, at 9:29 AM, Robert Casey <rob<at>iris.washington.edu <rob<at>iris.washington.edu>> wrote:
Hi MUSTANG users-
In addition to the aforementioned updates, we are to de-list the timing_drift metric from MUSTANG during the update process, as the IRIS Quality Assurance Advisory Committee agreed to the case made that it is currently ineffective in properly and consistently detecting timing drift at the indicated station. The QAAC will continue to revisit this metric to seek alternative algorithms that can better serve to make an appropriate measurement in this regard.
As always, we welcome feedback on our metrics and the algorithms used, and will continue to develop, enhance, and critique our metrics code to ensure they are functioning in the most effective way possible. In addition to this list, we also have a general discussion list for MUSTANG that you can sign up to by going to our Message Center: http://ds.iris.edu/message-center/topic/mustang-users/
-Rob
On Oct 24, 2016, at 4:24 PM, Robert Casey <rob<at>iris.washington.edu <rob<at>iris.washington.edu>> wrote:
Hi MUSTANG users-
Feature fixes and tests have been conducted for an important update to MUSTANG, specifically the web services that you use when getting access to the tabular (non-PSD) metrics. We are ready to put this into production tomorrow morning, October 25th, at 10 AM PDT. Problems are not anticipated with the switchover, and it may be entirely transparent to users, but I want to make sure everyone has a heads up just in case there is some disruption in service.
What you can expect in this new version should be to your liking. A summary of what is offered is the following:
Release version 1.2.0 (2016-10-25)
reworked code base to improve efficiencies
support for glob-style wildcards in all SNCL query terms
support for virtual network queries
support for ‘--’ location code queries
support for comma-separated query terms
support for ‘start’ and ‘end’ date-time terms
support for nodata term to send user specified HTTP code when no metrics are found
Improvements to helpful error message framework
LDDATE field is truncated to seconds instead of milli/microseconds
Allow partial matches for channel and location pairs found in certain metrics
Some metric values truncated to a specific precision
The on-line documentation for these features has been drafted for the 'measurements' service. The 'targets' and 'metrics' services will have their documents updated as well. Issues with the URL builder interface will be addressed as needed.
Thank you for your patience as we get this new install into production.
-Rob
-
-
