SA could do that, but could your report writer do that? As that seems to be
THE reason for the db, writing reports from it. 


Dirk Bulinckx. 
-----Original Message-----
From: Servers Alive Discussion List [mailto:[EMAIL PROTECTED] 
Sent: Thursday, May 11, 2006 6:15 PM
To: Servers Alive Discussion List
Subject: [SA-list] Re: [SA-list] RE: [SA-list] Re: [SA-list] RE: [SA-list]
Re: [SA-list] RE: [SA-list] Re: [SA-list] Enterprise level data storage (was
Skype Alerting)

WHy would have to be 1-1 or 1-many.  Why couldn't you just check each table
for the uid and consolidate (depending on table type).

Jason Passow
Mississippi Welders Supply
[EMAIL PROTECTED]
ph: (507) 494-5178
fax: (507) 454-8104

"If you do everything right, nobody will realize you've done anything at
all."



Servers Alive Discussion List wrote:
> That just the whole point.  We could indeed have a table for each 
> check type.  But how to link that in a proper way to a table with the host
info?
> The UID is indeed the column to use.  But for a report you need to 
> have a clear 1-to-1 or 1-to-many relationship (between the tables) and 
> in this example (with a table for each check type) you don't have that.
> You get a 1-to-(1 in one of these tables) relationship.
>
>  
>
>
> Dirk Bulinckx. 
> -----Original Message-----
> From: Servers Alive Discussion List [mailto:[EMAIL PROTECTED]
> Sent: Thursday, May 11, 2006 5:40 PM
> To: Servers Alive Discussion List
> Subject: [SA-list] Re: [SA-list] RE: [SA-list] Re: [SA-list] RE: 
> [SA-list]
> Re: [SA-list] Enterprise level data storage (was Skype Alerting)
>
> Perhaps I just didn't understand your last post, but why not just take 
> the text file and move it to a DB with appropriate columns and more 
> readable data (instead of 555555555 for schedules etc.)  Again I am 
> not going to say it is simple as I have not taken the time to figure 
> out the host file format.  Or perhaps several tables, one with the 
> hostname/ip, pretty name, and uid (maybe dependencies).  Then the 
> others can lead with uid and Check table, Alert table, and
Schedule/priority table.
>
> The alert table would have one record per alert (what, when, where, etc).
> The check table would have on record per check (I think) with type 
> criteria, timeout, second knock etc. (I guess this table could get a 
> bit complicated with different check types.)
>
> Just some thoughts.  I know nothing about the com Check but it seems 
> perhaps we would have to have a table for each type of com check (i.e.
> SMTP2POP3, ODBC, etc). It seems if it can all be put in flat file it 
> could all be put in a database.  Perhaps it would be as simple as 
> documenting the host file so that those who wish can put it in a DB.  
> I know others have figured out the structure.  Maybe the database 
> could simply export the flat file you require.  I wish I knew my way
around the host file a bit better.
>
>  From my ramblings above you can probably tell that I know very little 
> about the host file structure, and perhaps I am way off.  If so tell 
> me I have no idea and go on about your business.  If not then I am 
> happy to provide any input.
>
> Jason Passow
> Mississippi Welders Supply
> [EMAIL PROTECTED]
> ph: (507) 494-5178
> fax: (507) 454-8104
>
> "If you do everything right, nobody will realize you've done anything 
> at all."
>
>
>
> Servers Alive Discussion List wrote:
>   
>> We can put it in a DB by using about 5 fields.
>> Or by using about 5000 fields.  
>> Neither of the 2 would help you.
>>
>> Let me use a small example:
>>      ping host1
>>      The check itself has a couple of parameters:
>>              checktype (ping)
>>              number of frames to send
>>              % that has to respond
>>
>>
>>      url check <url>
>>      parameters:
>>              checktype (url http/https check)
>>              url
>>              get/post/head
>>              IF it's a POST then data_to_post
>>                      should contain/should not contain combined with this
>>     
> exact
>   
>> phrase/any of these words combined with <the_words>
>>                      or
>>                      CRC with "reset CRC after each check or not"
>>                      or
>>                      Site SSL Cert or CA SSL cert    combined with a
>> number of days
>>
>>              use authentication
>>                      if yes then USERNAME & PASSWORD
>>
>>              use proxy (and what proxy)
>>              HTTP version to use
>>
>>
>> An even worse example in an external check (COM based), here it's 
>> simple SA doesn't know what the syntax of the parameters are and as 
>> such not much data to show except something like "*NIX DISKSPACE 
>> CHECK", what mount and on what space we're checking, SA doesn't know 
>> (the
>>     
> COM knows, not SA).
>   
>> So what we could do is create a different table for each checktype 
>> (or maybe even SUB checktype) and then for each entry "link" to an 
>> entry in a
>>     
> table?
>   
>> Or we could have one large table with all different check options in it.
>> (could well be that we get to the limit of some DBS).
>>
>> I suppose this give a better view on why a DB isn't that simple....
>>
>>
>> Dirk Bulinckx. 
>> -----Original Message-----
>> From: Servers Alive Discussion List [mailto:[EMAIL PROTECTED]
>> Sent: Thursday, May 11, 2006 4:51 PM
>> To: Servers Alive Discussion List
>> Subject: [SA-list] Re: [SA-list] RE: [SA-list] Re: [SA-list] 
>> Enterprise level data storage (was Skype Alerting)
>>
>> Seems to me if you put it a database, their XML gurus could get the 
>> data and create their own XML.
>>
>> Jason Passow
>> Mississippi Welders Supply
>> [EMAIL PROTECTED]
>> ph: (507) 494-5178
>> fax: (507) 454-8104
>>
>> "If you do everything right, nobody will realize you've done anything 
>> at all."
>>
>>
>>
>> Servers Alive Discussion List wrote:
>>   
>>     
>>> "One thing that I do not think can be done (albeit I haven't tried 
>>> very
>>> hard) is putting the actual check properties into a report (aside from 
>>> HTML).   If the check details including , host name/ip, pretty name, 
>>> dependencies, alerts, check type,/criteria, schedule, etc were 
>>> stored in a database then I could create whatever reports I want"
>>> => those are indeed not in the DB.  But as XML seems to be what some 
>>> want, it can be done by creating an HTML template and have that 
>>> generate the XML format.
>>> What I would propose, is that someone on the list that wants that 
>>> XML, gives me an example of what should be in the XML file (make one 
>>> with 2
>>> entries) and I'll try to make a template for it.  And if changes are 
>>> needed to v6 (upcoming release that) for the template to be 
>>> possible, I'll make those changes.  How does that sound?
>>>
>>> Dirk Bulinckx. 
>>> -----Original Message-----
>>> From: Servers Alive Discussion List [mailto:[EMAIL PROTECTED]
>>> Sent: Thursday, May 11, 2006 3:31 PM
>>> To: Servers Alive Discussion List
>>> Subject: [SA-list] Re: [SA-list] Enterprise level data storage (was 
>>> Skype
>>> Alerting)
>>>
>>> I export the check results to a database each cycle and store it in 
>>> a database.  I can the pull up my web page and see the last 30 days
worth
>>> of checks using graphs etc.   I also have on that page a table showing 
>>> down cycles and length time before it recovered.  I could certainly 
>>> create the same report as an XML (if I knew anything about XML that
>>> is) or Excel (I know not an enterprise level App).  I would think I 
>>> could export the data and import into Crystal Reports or whatever 
>>> and
>>>     
>>>       
>> create a
>>   
>>     
>>> report.   Although I empathize with your struggle here I do not think 
>>> that Salive should be a reporting app.   
>>>
>>> One thing that I do not think can be done (albeit I haven't tried 
>>> very
>>> hard) is putting the actual check properties into a report (aside from 
>>> HTML).   If the check details including , host name/ip, pretty name, 
>>> dependencies, alerts, check type,/criteria, schedule, etc were 
>>> stored in a database then I could create whatever reports I want.
>>>
>>> Again I reiterate thank you Dirk for your patience and this kind of 
>>> continued dialog is unique and what makes SA far and away the best 
>>> app around.  Thank you for not stifling the discussion and allowing 
>>> ideas to flow.  On the other side of this discussion (maybe months 
>>> from now) I see some great things being added to SA.
>>>
>>>
>>> Jason Passow
>>> Mississippi Welders Supply
>>> [EMAIL PROTECTED]
>>> ph: (507) 494-5178
>>> fax: (507) 454-8104
>>>
>>> "If you do everything right, nobody will realize you've done 
>>> anything at all."
>>>
>>>
>>>
>>> Servers Alive Discussion List wrote:
>>>   
>>>     
>>>       
>>>> Your method of creating HTML on the fly is VERY unique when you 
>>>> look at all of the software in use in the enterprise environment. 
>>>> That means, that I (in the singular, me) have to code ALL reports, 
>>>> for all conceivable requests of my Sr. Management, and there is 
>>>> little ability to have a method of my up chain being able to ad-hoc 
>>>> browse through how tests are being done with out risk of injuring the
setup.
>>>> Also, have you tried to print an HTML page that's more than a page 
>>>> long? It looks like crud compared to the reports my managers are 
>>>> used to seeing from all of our other systems which typically use 
>>>> Crystal Reports for both text reporting and charting. On the other 
>>>> hand, if it was in a database, I can hand to any of a dozen people 
>>>> a table schema and say "write XYZ report" and actually have the 
>>>> report experts do their thing, rather than myself fumbling through it.
>>>>
>>>> A true enterprise solution has to have it's configuration data and 
>>>> historical data stored on a standardized data platform, preferably 
>>>> a DBMS such as MS SQL, DB2, Oracle, etc., so that the data can be 
>>>> protected and managed at the enterprise level through clustering, 
>>>> load balancing and Disaster Recovery techniques. This would mean 
>>>> that SA should not load any data internally or on the host PC 
>>>> running SA, but all would be in a DB server repository. Each check 
>>>> cycle would dynamically load each check's config from the DB 
>>>> server, and write back to the server the ongoing status of the check
cycle.
>>>> It sounds like this is going to be database intensive, but at the 
>>>> level I play at
>>>>     
>>>>       
>>>>         
>>> it's not.
>>>   
>>>     
>>>       
>>>> We have several systems that routinely write millions of data 
>>>> changes to our servers over the course of a 'light' day, and I 
>>>> suspect that's small compared to a lot of enterprises using a 
>>>> standard
>>>>         
> DBMS.
>   
>>>> Servers Alive and it's output is becoming incredibly tied to our 
>>>> contractual obligations to our customers, and it is becoming harder 
>>>> to defend a product developed at such a "personal" level. I can't 
>>>> WAIT until our customer charges us for server down time, and I 
>>>> can't prove him wrong because SA wiped out the uptime data during a 
>>>> server
>>>>       
>>>>         
>> re-boot.
>>   
>>     
>>>> You mostly see in the list where it's a dedicated individual who 
>>>> has chosen SA, but I also get the sense that it's hard for them to 
>>>> get others at their organization to work on it. Possibly because it 
>>>> is a unique product both in function and organizational concept.
>>>>
>>>> Don't get me wrong, I LOVE what you do, it's the how that I would 
>>>> like to see brought up to a true enterprise level.
>>>>
>>>> Michael D. Shook
>>>> Technical Analyst
>>>> Saddle Creek Corporation
>>>> [EMAIL PROTECTED]
>>>> 863 668 4477 (work)
>>>> 863 860 4070 (cell)
>>>> 863 665 1261 (fax)
>>>> www.saddlecrk.com
>>>>  
>>>>
>>>>   
>>>>     
>>>>       
>>>>         
>>>>> -----Original Message-----
>>>>> From: Servers Alive Discussion List [mailto:[EMAIL PROTECTED]
>>>>> Sent: Thursday, May 11, 2006 8:31 AM
>>>>> To: Servers Alive Discussion List
>>>>> Subject: [SA-list] RE: [SA-list] RE: [SA-list] RE: [SA-list]
>>>>> RE: [SA-list] RE: [SA-list] RE: [SA-list] RE: [SA-list] RE: 
>>>>> [SA-list] Skype alerting
>>>>>
>>>>> Is HTML a unique interface? 
>>>>> We can make reports in HTML.  You have to make an HTML template 
>>>>> for it, based on HTML tags (standard stuff for the layout part) 
>>>>> and specific tags for the actual data part.
>>>>> With whatever reportwriter it would be the same, it would need to 
>>>>> have access to the "format" and based on the format it could show 
>>>>> you what "vars" you can use.  We use tags.
>>>>>
>>>>>
>>>>> Dirk Bulinckx. 
>>>>>     
>>>>>       
>>>>>         
>>>>>           
>>>> --------------------------------------
>>>> The information contained in this message is intended only for the 
>>>> use of
>>>>     
>>>>       
>>>>         
>>> the addressee.  If the reader of this message is not the intended 
>>> recipient or agent of the intended recipient, you are hereby 
>>> notified that any dissemination, distribution, or copying of the 
>>> message is strictly prohibited.
>>>   
>>>     
>>>       
>>>> To unsubscribe send a message with UNSUBSCRIBE as subject to
>>>>     
>>>>       
>>>>         
>>> [email protected]
>>>   
>>>     
>>>       
>>>>     
>>>>       
>>>>         
>>> To unsubscribe send a message with UNSUBSCRIBE as subject to 
>>> [email protected] To unsubscribe send a message with UNSUBSCRIBE 
>>> as subject to [email protected]
>>>
>>>
>>>     
>>>       
>> To unsubscribe send a message with UNSUBSCRIBE as subject to 
>> [email protected] To unsubscribe send a message with UNSUBSCRIBE as 
>> subject to [email protected]
>>
>>
>>     
> To unsubscribe send a message with UNSUBSCRIBE as subject to 
> [email protected] To unsubscribe send a message with UNSUBSCRIBE as 
> subject to [email protected]
>
>
To unsubscribe send a message with UNSUBSCRIBE as subject to
[email protected]
To unsubscribe send a message with UNSUBSCRIBE as subject to [email protected]

Reply via email to