Service Issues

Report any Haven OnDemand API service issues to the development team here. Before submitting your issue and for issue status definitions, be sure to read this page.
Search the Dev Community
Showing results for 
Search instead for 
Do you mean 
Post a Message
0 Likes
Status:
Completed

RecognizeSpeech Error

Status:
Completed
by Gordo Level 4 on ‎04-27-2015 04:11

I uploaded a WMA file to the service for transcription :

curl "https://api.idolondemand.com/1/api/async/recognizespeech/v1" -X POST -H Content-Type:multipart/form-data; --insecure -F apikey=<APIKEY> -F language=enuk -F file=@target_london_811_0186.WMA


{
  "jobID": "usw3p_c78ce6e7-5e95-4fdc-afab-23ef3c7408f2"
}

 After a wihle I check the status:

 

curl "https://api.idolondemand.com/1/job/status/usw3p_c78ce6e7-5e95-4fdc-afab-23ef3c7408f2" --insecure -F apikey=<APIKEY>

 The response

 

 

{
  "actions": [
    {
      "errors": [
        {
          "error": 5007,
          "reason": "Key Management Service Error"
        }
      ],
      "status": "failed",
      "action": "recognizespeech",
      "version": "v1"
    }
  ],
  "jobID": "usw3p_c78ce6e7-5e95-4fdc-afab-23ef3c7408f2",
  "status": "failed"
}

 Why does this happen?

 

Status:
Completed
0 Likes
Status:
Completed

Hi

 

I am using the HTTP connector to index a site. I see that lot of data is not relevant and i want to remove it.

 

I tried the OOTB feaure Pre0=HtmlExtraction in CFS.cfg however i still see some data. Is it possible to custimize this OOTB HTML Extraction?

 

OR  

 

Do I need to write a LUA script to remove all the irelevant content? Is it possible to share such a LUA script with regex?

 

I am using IDOL 10.0 for now and will be updating it to 10.9 soon.

 

 

Thanks

Status:
Completed

Hi @VabsSharma,

 

Our recent update added clipping functionality to our Web Connector. In particular the clip page parameter available is documented in the link below.

 

https://dev.havenondemand.com/docs/Connectors_Web.html

 

Regards,

 

Jenn

Haven OnDemand team

0 Likes
Status:
Completed

Speech server caching results

Status:
Completed
by djpidol Level 3 on ‎05-14-2015 07:44

Speech server appears to be delivering the same result for different target URLs, I have checked the request log to verify that my code is sending different URL mp3

source code https://github.com/djpidol/bearded-octo-sansa/blob/master/thought_for_the_day.rb

Status:
Completed

A number of updates to the Speech Recognition API have been deployed recently and this issue should now be resolved.


Regards


Chris
Haven OnDemand team

0 Likes
Status:
Completed

HTTPS issue

Status:
Completed
by Ward Level 3 on ‎02-24-2015 06:54

I'm seeing the following error when I try to connect to IDOL OnDemand since about 4 PM EST today:

 

Creating index merge_case_index1 failed with error java.lang.RuntimeException: Could not generate DH keypair
 
There another user who is trying to connect from Salesforce.com to IDOL who is experiencing the same issue.  Seems like maybe there has been a certificate expiration or something.  Is that possible?
Status:
Completed

Ward,

 

This issue has be traced to the continued usage of Java 6 which is unable to negotiate SSL to modern standards.  As an interim measure until Salesforce update to a later version , we are permitting lower grade ciphers for connections from the Salesforce IP range.

 

Regards


Chris

0 Likes
Status:
Completed

Speed Issues

Status:
Completed
by Gordo Level 4 on ‎02-20-2015 12:59

Hi,

 

We are consistently noticing slow query speed compared to what we'd expect from a standard IDOL.

 

The below timings are taken from the code and indiciate the time taken to make and receive the response from IOD (we notice similar speeds when using the Find sample application). This makes demoing the platform problematic . 

 

The below queries took place at 2015-02-19 15:12

 

https ://api.idolondemand.com/1/api/sync/findrelatedconcepts/v1?text=pensions&field_text=&start=1&print=no_results&indexes=news_eng&&max_page_results=100&absolute_max_results=100&querysummary=true&apikey=<APIKEY>
1568 ms to execute query

 

https: //api.idolondemand.com/1/api/sync/getparametricvalues/v1?text=pensions&indexes=news_eng&&field_text=&sort=document_count&document_count=true&field_name=RSS_SOURCE,RSS_CATEGORY,WIKIPEDIA_TYPE,PERSON_PROFESSION,PLACE_COUNTRY_CODE,COMPANY_EXCHANGE&max_values=20&apikey=<APIKEY>
1275 ms to execute query

 

https: //api.idolondemand.com/1/api/sync/querytextindex/v1?text=pensions&max_page_results=10&absolute_max_results=10&indexes=news_eng&&field_text=&summary=concept&print_fields=autn:title,DREDBNAME,DATE,RSS_URL,RSS_RANK,RSS_COPYRIGHT,RSS_SOURCE,RSS_CATEGORY,MODIFIED_DATE,WIKIPEDIA_TYPE,WIKIPEDIA_ALIAS,WIKIPEDIA_CATEGORY,WIKIPEDIA_INFOBOX,WIKIPEDIA_ALTTITLE,WIKIPEDIA_TITLE_ENG,CONTENT_LENGTH,WIKIPEDIA_RANK,WIKIPEDIA_ID,WIKIDATA,PERSON_PROFESSION,PERSON_DATE_OF_BIRTH,PERSON_DATE_OF_DEATH,LON_NUMBER,LAT_NUMBER,PLACE_POPULATION,PLACE_COUNTRY_CODE,PLACE_REGION1,PLACE_REGION2,PLACE_ELEVATION,COMPANY_RIC,COMPANY_TICKER,COMPANY_EXCHANGE,COMPANY_WEBSITE&start=1&total_results=true&sort=relevance&highlight=terms&sentences=3&predict=false&start_tag=<span class=resulthighlight>&apikey=<APIKEY>
2857 ms to execute query

 

https: //api.idolondemand.com/1/api/sync/querytextindex/v1?text=pensions&max_page_results=10&absolute_max_results=10&indexes=news_eng&&field_text=&summary=concept&print=no_results&start=1&total_results=true&sort=relevance&highlight=terms&sentences=3&predict=false&start_tag=<span class=resulthighlight>&apikey=<APIKEY>
3320 ms to execute query

 

We'd at least expect sub-second response times for these simple queries.

 

Thanks

 

Gordon

 

Status:
Completed
0 Likes
Status:
Completed

I'm not sure if this is the same issue as  IODI-166 but there is a disparity between counts made in the calls to getparametricvalues and querytextindex

 

URL:

 

https //api.idolondemand.com/1/api/sync/getparametricvalues/v1?text=mars&indexes=news_eng&&field_text=( MATCH{the%20telegraph}:*/RSS_SOURCE )&sort=document_count&document_count=true&field_name=RSS_SOURCE,RSS_CATEGORY,WIKIPEDIA_TYPE,PERSON_PROFESSION,PLACE_COUNTRY_CODE,COMPANY_EXCHANGE&max_values=20&apikey=<APIKEY>

 Results

  

"RSS_SOURCE":
{
"THE TELEGRAPH": 998
}

 

URL : 

 

https //api.idolondemand.com/1/api/sync/querytextindex/v1?text=mars&max_page_results=10&absolute_max_results=10&indexes=news_eng&&field_text=( MATCH{the%20telegraph}:*/RSS_SOURCE )&summary=concept&print_fields=autn:title,DREDBNAME,DATE,RSS_URL,RSS_RANK,RSS_COPYRIGHT,RSS_SOURCE,RSS_CATEGORY,MODIFIED_DATE,WIKIPEDIA_TYPE,WIKIPEDIA_ALIAS,WIKIPEDIA_CATEGORY,WIKIPEDIA_INFOBOX,WIKIPEDIA_ALTTITLE,WIKIPEDIA_TITLE_ENG,CONTENT_LENGTH,WIKIPEDIA_RANK,WIKIPEDIA_ID,WIKIDATA,PERSON_PROFESSION,PERSON_DATE_OF_BIRTH,PERSON_DATE_OF_DEATH,LON_NUMBER,LAT_NUMBER,PLACE_POPULATION,PLACE_COUNTRY_CODE,PLACE_REGION1,PLACE_REGION2,PLACE_ELEVATION,COMPANY_RIC,COMPANY_TICKER,COMPANY_EXCHANGE,COMPANY_WEBSITE&start=1&total_results=true&sort=relevance&highlight=terms&sentences=3&predict=false&start_tag=<span class=resulthighlight>&apikey=<APIKEY>

 Results:

 

"totalhits": 958

 Why is there a difference? Is it just down to predicting the values? If so, is there a way to disable predicts?

 

Thanks

 

Gordon

 

 

 

Status:
Completed
0 Likes
Status:
Completed

Broken Parametric Refinement

Status:
Completed
by Gordo Level 4 on ‎02-18-2015 05:34

https://api.idolondemand.com/1/api/sync/querytextindex/v1?text=pensions&start_tag=<span class=result...

 

This gives the error 

 

{
  "error": 4024,
  "reason": "Job actions failed",
  "actions": [
    {
      "errors": [
        {
          "error": 4024,
          "reason": "Field_text contains operations on unoptimized fields",
          "detail": "'The fieldtext contained an operation against a non-optimized field.'"
        }
      ],
      "status": "failed",
      "action": "querytextindex",
      "version": "v1"
    }
  ],
  "status": "failed"
}

 

According to the documentation :

 

https://www.idolondemand.com/developer/docs/PublicDatasets.html

 

rss_sourceStringParametric

 

This worked last week.

Status:
Completed
0 Likes
Status:
In Progress

Patents Date Restriction

Status:
In Progress
by Gordo Level 4 on ‎03-24-2015 05:12

Hi, I'm trying to do some date restrictions on the patents corpus:

 

curl "https://api.idolondemand.com/1/api/sync/querytextindex/v1?text=DREALL&field_text=RANGE%7B1364125215e%2C.%7D%3Apatent_pub_date+AND+RANGE%7B.%2C1427197215e%7D%3Apatent_pub_date&indexes=patents&apikey=<APIKEY>"

 

This returns an error  

 

{
  "error": 4024,
  "reason": "Job actions failed",
  "actions": [
    {
      "errors": [
        {
          "error": 4024,
          "reason": "Field_text contains operations on unoptimized fields",
          "detail": "'The fieldtext contained an operation against a non-optimized field.'"
        }
      ],
      "status": "failed",
      "action": "querytextindex",
      "version": "v1"
    }
  ],
  "status": "failed"
}

 

Is the patent index not setup to handle this?

 

Thanks

 

Gordon

 

Status:
In Progress

Gordo,


Thanks for taking the time to raise this issue.

 

The Patent index is not currently configured for data operations, however this is intended.  Our engineers will be making the appropriate changes and we will let you know when this is complete.

 

Regards

 

Chris

IDOL OnDemand team

0 Likes
Status:
Completed

Create Index Problem

Status:
Completed
by Gordo Level 4 on ‎05-06-2015 01:46

I'm trying to create two new custom-field indexes on IoD.

 

Apparently I have 0/50 units used (though I have 1 index with custom fields)

A new custom index costs 12 units

When I try and add a new index the message says 

 

"Unable to create text index: Reached index quota - need 12 units, have 8 units left"

 

Can you explain why this is?

 

Thanks

 

Gordon

 

Status:
Completed

Hi Gordo,

 

The incorrect display of text index quotas you identified as been removed.

 

You can view you quotas at any time via: https://www.havenondemand.com/account/quotas.html

 

Regards


Chris

Haven OnDemand team

 

 

0 Likes
Status:
Not an Issue

Patents QueryTextIndex by reference

Status:
Not an Issue
by Gordo Level 4 on ‎03-23-2015 09:22

If I do the following query:

 

curl "https://api.idolondemand.com/1/api/sync/querytextindex/v1?text=DREALL&indexes=patents&summary=concept&apikey=<APIKEY>"

 I get back some results:

 

 "documents": [
    {
      "reference": "http://www.google.com/patents/USD07640597",
      "weight": 85.35,
      "links": [],
      "index": "patents",
      "title": "Image display device",
.....

 If i then try and query for this reference:

 

curl "https://api.idolondemand.com/1/api/sync/querytextindex/v1?reference=http%3A%2F%2Fwww.google.com%2Fpatents%2FUSD07640597&indexes=patents&apikey=<APIKEY>"

 I get the following:

 

{
  "error": 5013,
  "reason": "Job actions failed",
  "actions": [
    {
      "errors": [
        {
          "error": 5013,
          "reason": "Reference does not exist"
        }
      ],
      "status": "failed",
      "action": "querytextindex",
      "version": "v1"
    }
  ],
  "status": "failed"
}

 Is this a bug or am I misunderstanding something?

Status:
Not an Issue

Gordo,

 

Reference is his case is a reference to a file uploaded via the storeobject API rather than an index reference.

 

If you are trying to retrieve the given document you will need to use getcontent & the index_reference parameter, or if you would like to do findsimilar to find similar documents, you will again need to pass the reference as index_reference rather than reference.

 

Regards


Chris

IDOL OnDemand team

0 Likes
Status:
Completed

Arxiv GetContent

Status:
Completed
by Gordo Level 4 on ‎03-20-2015 07:31

If we try and do a GetContent request for the Arxiv index we get an error :

 

 

curl "http://<REMOVEME>api.idolondemand.com/1/api/sync/getcontent/v1?indexes=arxiv&index_reference=http%3A%2F%2Farxiv.org%2...

 

{
  "error": 8002,
  "reason": "Job actions failed",
  "actions": [
    {
      "errors": [
        {
          "error": 8002,
          "reason": "Index name invalid"
        }
      ],
      "status": "failed",
      "action": "getcontent",
      "version": "v1"
    }
  ],
  "status": "failed"
}
Status:
Completed

Thanks Gordo for taking time to raise this issue.

 

Our Engineers have now enabled the Arxiv index for getcontent.  Please do let us know if you have any further issues.

 

Regards

 

Chris

IDOL OnDemand Team

0 Likes
Status:
Open

Text Index not responding or timeout

Status:
Open
by Jorge New Member ‎07-01-2016 08:51 - edited ‎07-01-2016 08:52

Hello,

 

Today (1 july) Text Index is not responding. Is it me or server has intermittent service? Thanks for your help.

 

Jorge

0 Likes
Status:
Not an Issue

Speech recognition error 5000

Status:
Not an Issue
by krishnaguy New Member on ‎03-12-2016 04:17

I am trying to call the speech to text services via Python, but get a "Backend request failed" error. However I do get results when I upload the file via the "Try it" interface.

 

My code snippet below:

 

import requests

r = requests.post('https://api.havenondemand.com/1/api/async/recognizespeech/v1', data = {'apikey':'KEY', 'language':'en-US-tel', 'file':'file_name_with_full_path'})

 

and the error:

 

{u'actions': [{u'action': u'recognizespeech',
   u'errors': [{u'error': 5000, u'reason': u'Backend request failed'}],
   u'status': u'failed',
   u'version': u'v1'}],
 u'jobID': u'JOB-ID',
 u'status': u'failed'}

 

I have attached the sample wav file that I was using as well. 

Status:
Not an Issue

Hi Krishnaguy,


Thanks for getting back to us with the update.


Do let us know if you need anything further.

 

Chris
Haven OnDemand team

0 Likes
Status:
Open

When I perform a synchronous call to the sentiment analysis API via the HOD Python module and the first non-space character in the input text is & or #, the output is None. 

0 Likes
Status:
Completed

View Document - "Error 5000"

Status:
Completed
by GuillermoGomez New Member on ‎03-07-2016 09:44

Hi,

 

The view document feature was working fine for me about a week ago in my project.  Then I suddenly started recieving errors any time I make a request to the document viewer.  Other parts of the API work just fine, such as the text extraction.  

 

When testing on the haven on demand website I recieve the same errors (Tried this for both .docx and .txt files):

{
  "error": 5000,
  "reason": "Job actions failed",
  "actions": [
    {
      "errors": [
        {
          "error": 5000,
          "reason": "Backend request failed",
          "detail": "Worker request failed"
        }
      ],
      "status": "failed",
      "action": "viewdocument",
      "version": "v1"
    }
  ],
  "status": "failed"
}

 

All the files I have tried so far do contain content and I am not just passing it an empty file.

 

Any suggestions to fixing this problem is greatly appreciated, thanks in advance!

Status:
Completed
0 Likes
Status:
Open

Hi,

 

I have tried to use your "Contact Us" form, from both Chrome (v53) and Edge on a Windows 10 machine, but the submit button does nothing. I've tried it both logged in and logged out of your site.

 

 

0 Likes
Status:
Open

Key Management Service Error 5100

Status:
Open
by bcormode New Member on ‎04-28-2016 01:49

I'm trying to create a Standard text index using the 'Create Text Index' on the Try It interface.

 

I was able to create and delete a Text Index earlier today (a couple of times), but now it's not working.

 

I'm getting the following error:

 

{
  "error": 5100,
  "reason": "Job actions failed",
  "actions": [
    {
      "errors": [
        {
          "error": 5100,
          "reason": "Key Management Service Error",
          "detail": "<html><head><title>Apache Tomcat/7.0.52 (Ubuntu) - Error report</title><style><!--H1 {font-family:Tahoma,Arial,sans-serif;color:white;background-color:#525D76;font-size:22px;} H2 {font-family:Tahoma,Arial,sans-serif;color:white;background-color:#525D76;font-size:16px;} H3 {font-family:Tahoma,Arial,sans-serif;color:white;background-color:#525D76;font-size:14px;} BODY {font-family:Tahoma,Arial,sans-serif;color:black;background-color:white;} B {font-family:Tahoma,Arial,sans-serif;color:white;background-color:#525D76;} P {font-family:Tahoma,Arial,sans-serif;background:white;color:black;font-size:12px;}A {color : black;}A.name {color : black;}HR {color : #525D76;}--></style> </head><body><h1>HTTP Status 500 - Internal Server Error</h1><HR size=\"1\" noshade=\"noshade\"><p><b>type</b> Status report</p><p><b>message</b> <u>Internal Server Error</u></p><p><b>description</b> <u>The server encountered an internal error that prevented it from fulfilling this request.</u></p><HR size=\"1\" noshade=\"noshade\"><h3>Apache Tomcat/7.0.52 (Ubuntu)</h3></body></html>"
        }
      ],
      "status": "failed",
      "action": "createtextindex",
      "version": "v1"
    }
  ],
  "status": "failed"
}

I checked my Account Subscription Quotas and they seem to be fine. 

0 Likes
Status:
Not an Issue

500 - Backend Request Failed

Status:
Not an Issue
by justing Level 2 on ‎12-05-2015 12:43

Hi,

 

While attempting to use the view document API, I've recently started hitting a 500 internal error - here's the JSON

 

{
    "error": 5000,
    "reason": "Job actions failed",
    "actions": [
        {
            "errors": [
                {
                    "error": 5000,
                    "reason": "Backend request failed",
                    "detail": "Worker request failed"
                }
            ],
            "status": "failed",
            "action": "viewdocument",
            "version": "v1"
        }
    ],
    "status": "failed"
}

Is this a configuration issue on our part? 

0 Likes
Status:
Completed

HavenonDemand migration

Status:
Completed
by djpidol Level 3 on ‎10-12-2015 07:20

Hi, I've updated a script to send requests to the new havenondemand api. STT appears to be taking longer than would be expected. I checked the request log for my key, however, this shows no requests in last 24 hours. When will the transition be completed? thanks

Status:
Completed

Hi,

 

The transition to Haven OnDemand is now complete and you should be able to see your requests in the request log.

 

Do let us know if you have any remaining issues or feedback.


Regards

 

Chris

Haven OnDemand team

0 Likes
Status:
In Progress

Hello, I have several API keys that were compromised, and I would like to delete them as soon as possible, but there seems to be no option to delete the API keys in the account settings. Thank you.

Status:
In Progress

Thanks Johnsonshi,

 

The feature you mention is on our roadmap for later in the year.  In the meantime if you would like us to delete any API Keys for you - please provide details via the contact us form. (https://www.havenondemand.com/contactus.html)


Regards

 

Chris

Haven OnDemand team

Social Media
† The opinions expressed above are the personal opinions of the authors, not of HPE. By using this site, you accept the Terms of Use and Rules of Participation