[analog-help] dns lookup hostnames
Not sure if this is just a documentation problem, or what. The hostexclude documentation says: HOSTINCLUDE 131.111.20.18 # simple IP address HOSTINCLUDE 131.111.20.* # wildcard HOSTINCLUDE 131.111.20 # the same meaning HOSTINCLUDE 131.111.20-23 # a range of class C addresses HOSTINCLUDE 131.111.20.18/23 # subnet mask I've got hostexclude 100.100.111 to exclude internal addresses (allocated via dhcp or similar). Naturally some of these fail to look up via analog's dnslookup (days or weeks later). Having been inserted into the dnslookup file with a * (not found) they appear on the host (and other) reports as 100.100.111.24 or similar: the hostexclude doesn't eliminate them. Changing it to hostexclude 100.100.111.* fixes the problem. Caused me some concern when I thought external hosts were accessing admin stuff that they should have been blocked from ! I also have hostexclude 100.100.100-101 - not sure what the ideal syntax would be there, but they're not a big problem as they're static addresses. Regards Chris Any views expressed in this message are those of the sender and not necessarily those of CCA Group. The unauthorized use, disclosure, copying or alteration of this message is forbidden. The contents of this message may be confidential and/or privileged, copyright CCA Group and are intended solely for the use of the individual or entity to whom they are addressed. Whilst this message has been scanned, CCA Group cannot guarantee that it is virus free or compatible with your systems and accepts no responsibility for any loss or damage arising from its use. The recipient is advised to run their own anti-virus software. If you receive this message in error please contact [EMAIL PROTECTED] immediately, destroy any copies and delete it from your computer systems. + | TO UNSUBSCRIBE from this list: |http://lists.meer.net/mailman/listinfo/analog-help | | Analog Documentation: http://analog.cx/docs/Readme.html | List archives: http://www.analog.cx/docs/mailing.html#listarchives | Usenet version: news://news.gmane.org/gmane.comp.web.analog.general +
Re: [analog-help] howto change the filename of the output images?
,analog-help@lists.meer.net wrote: hey all, I'm trying to change the filename of the output chartfiles to a different filename. The default filename for the referrers is refsite.png. I would like to change this to %mrefsite.png. Is this possible? I really don't want to use different directories per file Yes. Just add a trailing %m to your chartdir localchartdir They specify a prefix, rather than a directory. Chris Any views expressed in this message are those of the sender and not necessarily those of CCA Group. The unauthorized use, disclosure, copying or alteration of this message is forbidden. The contents of this message may be confidential and/or privileged, copyright CCA Group and are intended solely for the use of the individual or entity to whom they are addressed. Whilst this message has been scanned, CCA Group cannot guarantee that it is virus free or compatible with your systems and accepts no responsibility for any loss or damage arising from its use. The recipient is advised to run their own anti-virus software. If you receive this message in error please contact [EMAIL PROTECTED] immediately, destroy any copies and delete it from your computer systems. + | TO UNSUBSCRIBE from this list: |http://lists.meer.net/mailman/listinfo/analog-help | | Analog Documentation: http://analog.cx/docs/Readme.html | List archives: http://www.analog.cx/docs/mailing.html#listarchives | Usenet version: news://news.gmane.org/gmane.comp.web.analog.general +
Re: [analog-help] Question re. capturing request data on internal anchor links for a particular page
Support for analog web log analyzer wrote: Hi… The problem I am having is this: On a site about animals there is a page called pets.asp The page contains a number of anchors which look like this: /pets.asp#dogs /pets.asp#cats /pets.asp#turtles I need to capture all the requests for each anchored section of the site individually. Anchors are never sent to the server - if you go from within the page, there is no server request, it's just handled by the browser locally. If you come in from outside, the server is asked for pets.asp, and then the browser finds #dog. The only way to get the server involved would be to use /pets.asp?dogs#dogs or similar - then you'd get it logged, and analog could analyse it. Chris Any views expressed in this message are those of the sender and not necessarily those of CCA Group. The unauthorized use, disclosure, copying or alteration of this message is forbidden. The contents of this message may be confidential and/or privileged, copyright CCA Group and are intended solely for the use of the individual or entity to whom they are addressed. Whilst this message has been scanned, CCA Group cannot guarantee that it is virus free or compatible with your systems and accepts no responsibility for any loss or damage arising from its use. The recipient is advised to run their own anti-virus software. If you receive this message in error please contact [EMAIL PROTECTED] immediately, destroy any copies and delete it from your computer systems. + | TO UNSUBSCRIBE from this list: |http://lists.meer.net/mailman/listinfo/analog-help | | Analog Documentation: http://analog.cx/docs/Readme.html | List archives: http://www.analog.cx/docs/mailing.html#listarchives | Usenet version: news://news.gmane.org/gmane.comp.web.analog.general +
Re: [analog-help] Question re. capturing request data oninternalanchor links for a particular page
Support for analog web log analyzer wrote: Yeah, thanks Chris...Helpful...Presumably you could just re-write those links in the code to get it to play ball? Yes. But be aware that if you did that to internal links (ie within page.asp to page.asp) you'd probably be forcing a reload - instead of instantly refreshing to the right point, the user would see a page reload, then a move to the right point, and your server would have to do the extra work. External links would force caches to treat pets.asp?dogs as a different page to page.asp?cats again probably increasing workload and response time. Chris Chris Sharman [EMAIL PROTECTED] wrote: The only way to get the server involved would be to use /pets.asp?dogs#dogs or similar - then you'd get it logged, and analog could analyse it. Any views expressed in this message are those of the sender and not necessarily those of CCA Group. The unauthorized use, disclosure, copying or alteration of this message is forbidden. The contents of this message may be confidential and/or privileged, copyright CCA Group and are intended solely for the use of the individual or entity to whom they are addressed. Whilst this message has been scanned, CCA Group cannot guarantee that it is virus free or compatible with your systems and accepts no responsibility for any loss or damage arising from its use. The recipient is advised to run their own anti-virus software. If you receive this message in error please contact [EMAIL PROTECTED] immediately, destroy any copies and delete it from your computer systems. + | TO UNSUBSCRIBE from this list: |http://lists.meer.net/mailman/listinfo/analog-help | | Analog Documentation: http://analog.cx/docs/Readme.html | List archives: http://www.analog.cx/docs/mailing.html#listarchives | Usenet version: news://news.gmane.org/gmane.comp.web.analog.general +
Re: [analog-help] Report on portion of querystring
Support for analog web log analyzer wrote: We use Oracle reports for reporting and the url looks like this: /dev60cgi/rwcgi60.exe?chk_bysubject_smallPDFAND%20Aetc How can I make it group on the first parameter (chk_bysubject_small which is the report name) so I can determine which reports are being called? I've been reading about FILEALIAS, but I can't seem to get it to do what I want. I've used FILEALIAS to move bits of the querystring into the path. Eg FILEALIAS /dev60cgi/rwcgi60.exe?** /dev60cgi/rwcgi60.exe-$1?$2 This works for me, but Stephen's suggestion of internal search reports sounds interesting - it would enable you to pick out values of named parameters. I don't see, from a quick read, whether/how it picks out positional parameters ? Chris Any views expressed in this message are those of the sender and not necessarily those of CCA Group. The unauthorized use, disclosure, copying or alteration of this message is forbidden. The contents of this message may be confidential and/or privileged, copyright CCA Group and are intended solely for the use of the individual or entity to whom they are addressed. Whilst this message has been scanned, CCA Group cannot guarantee that it is virus free or compatible with your systems and accepts no responsibility for any loss or damage arising from its use. The recipient is advised to run their own anti-virus software. If you receive this message in error please contact [EMAIL PROTECTED] immediately, destroy any copies and delete it from your computer systems. + | TO UNSUBSCRIBE from this list: |http://lists.meer.net/mailman/listinfo/analog-help | | Analog Documentation: http://analog.cx/docs/Readme.html | List archives: http://www.analog.cx/docs/mailing.html#listarchives | Usenet version: news://news.gmane.org/gmane.comp.web.analog.general +
Re: [analog-help] Idiot's Guide for IIS
Support for analog web log analyzer wrote: They're lines that don't have a Status code. (Analog creates 2 logformats when it autoanalyzes a W3C log - one matches the #Fields line, and the other is just (#%j), to handle the other 3 header lines. These lines obviously don't have a %c field. You have 84 sets of 4 header lines (336 in total), and 84 sets of 3 lines that match (#%j), qhixh ia 252 lines without a status code. Ah, ok, so Analog doesn't really understand comments, just treats them as another data line (and warns) - now I get it. Unwanted logfile entries: 792 -- I thought these were unwanted dates ? Unwanted lines are any lines that you have EXCLUDEd, or no INCLUDEd, or that fall outside a specified FROM-TO period. I didn't think I'd excluded any, but I'd forgotten the HOSTEXCLUDEs in my standard config file (to exclude insiders). Oops. IIS, even on Server 2003, does not log byte counts or referrers by default. You will need to go into your IIS manager, select the properties for the web site and click Properties... in the Enable logging section of the Web Site tab of the dialog box. In the Extended Properties tab of the new dialog, you can select which fields are recorded. I would add bytes and referrers, but nothing more. I'll pass that along. Many Thanks Chris Any views expressed in this message are those of the sender and not necessarily those of CCA Group. The unauthorized use, disclosure, copying or alteration of this message is forbidden. The contents of this message may be confidential and/or privileged, copyright CCA Group and are intended solely for the use of the individual or entity to whom they are addressed. Whilst this message has been scanned, CCA Group cannot guarantee that it is virus free or compatible with your systems and accepts no responsibility for any loss or damage arising from its use. The recipient is advised to run their own anti-virus software. If you receive this message in error please contact [EMAIL PROTECTED] immediately, destroy any copies and delete it from your computer systems. + | TO UNSUBSCRIBE from this list: |http://lists.meer.net/mailman/listinfo/analog-help | | Analog Documentation: http://analog.cx/docs/Readme.html | List archives: http://www.analog.cx/docs/mailing.html#listarchives | Usenet version: news://news.gmane.org/gmane.comp.web.analog.general +
[analog-help] Idiot's Guide for IIS
For something truly terrible I did in a past life, I now have stats from an IIS server to analyze. Am I right in thinking that it's just telling me (warnings below) that neither bytes nor referrer are being logged by IIS ? Logfile lines without status code: 252 -- what are these ? Unwanted logfile entries: 792 -- I thought these were unwanted dates ? I've got 336 lines beginning #, all the rest begin 2005-10-0. I asked for ( got) all of October (so far). Does IIS log referrers separately ? Thanks Chris Warning M: Logfile BATCH:[WEBANAL.QBCONN]EX051009.LOG;1 contains lines with no bytes: byte counts may be low Warning R: Turning off empty Referrer Report Warning R: Turning off empty Failed Referrer Report Warning R: Turning off empty Search Query Report Warning R: Turning off empty Search Word Report Warning R: Turning off empty File Size Report Warning R: In Domain Report, turning off pie chart of only one wedge Warning R: In File Type Report, turning off empty pie chart Warning R: In Directory Report, turning off empty pie chart #Software: Microsoft Internet Information Services 6.0 #Version: 1.0 #Date: 2005-10-06 10:45:56 #Fields: date time s-ip cs-method cs-uri-stem cs-uri-query s-port cs-username c-ip cs(User-Agent) sc-status sc-substatus sc-win32-st atus Any views expressed in this message are those of the sender and not necessarily those of CCA Group. The unauthorized use, disclosure, copying or alteration of this message is forbidden. The contents of this message may be confidential and/or privileged, copyright CCA Group and are intended solely for the use of the individual or entity to whom they are addressed. Whilst this message has been scanned, CCA Group cannot guarantee that it is virus free or compatible with your systems and accepts no responsibility for any loss or damage arising from its use. The recipient is advised to run their own anti-virus software. If you receive this message in error please contact [EMAIL PROTECTED] immediately, destroy any copies and delete it from your computer systems. + | TO UNSUBSCRIBE from this list: |http://lists.meer.net/mailman/listinfo/analog-help | | Analog Documentation: http://analog.cx/docs/Readme.html | List archives: http://www.analog.cx/docs/mailing.html#listarchives | Usenet version: news://news.gmane.org/gmane.comp.web.analog.general +
Re: [analog-help] Date based Log File Analysis.
Would it be of help, if we could Enhance the current version to include A date based processing. I mean, that the end user can analyse the Log-files between x to y dates. Yes, this is an enhancement I've suggested before. I have an approach, to achieve this, 1. Write a daemon that will continously monitor the log-file specified in the .cfg file and tar the logs if a certain chosen capacity is exceeded. In this way keep taring the files. Don't understand the relevance of this step ? 2. once Analog is executed (with the dates selected), execute a script analog.pl to work on the tar files to extract and recreate the file for the specific time period by doing some header/trailer processing. Sounds like a fairly intensive approach, particularly for large gzipped logs. Sean said: Is this not already possible quite easily using the FROM, TO and LOGFILE statements? No - if you have multiple logfiles, only the TO date is used to generate the names to read. If you have monthly logs, and want a 3-month report, you can either set up your LOGFILE statement to read all logs ( discard unwanted months), or read a year's logs (discarding unwanted), or you'll only get one month. None of these are really desireable. The correct approach would be for analog to loop round generating logfile names from FROM to TO ... It doesn't seem difficult, but I haven't a pressing enough need to put the time in myself - for me it's more of a gotcha than a real problem. Chris Any views expressed in this message are those of the sender and not necessarily those of CCA Group. The unauthorized use, disclosure, copying or alteration of this message is forbidden. The contents of this message may be confidential and/or privileged, copyright CCA Group and are intended solely for the use of the individual or entity to whom they are addressed. Whilst this message has been scanned, CCA Group cannot guarantee that it is virus free or compatible with your systems and accepts no responsibility for any loss or damage arising from its use. The recipient is advised to run their own anti-virus software. If you receive this message in error please contact [EMAIL PROTECTED] immediately, destroy any copies and delete it from your computer systems. + | TO UNSUBSCRIBE from this list: |http://lists.meer.net/mailman/listinfo/analog-help | | Analog Documentation: http://analog.cx/docs/Readme.html | List archives: http://www.analog.cx/docs/mailing.html#listarchives | Usenet version: news://news.gmane.org/gmane.comp.web.analog.general +
Re: [analog-help] Date based Log File Analysis.
Support for analog web log analyzer wrote: Is this not already possible quite easily using the FROM, TO and LOGFILE statements? Yeah. Thanks. But, How do we make sure that all the files are Saved for analysis in a future date ?? Now I understand why you were talking about splitting and tarring. Most webservers (including Apache OSU which I use) already have some kind of facility or add-on for what's usually called 'log file rotation'. Typically you choose some manageable interval (month, day, or whatever), and have the server open a new log, and have some date-based naming scheme - perhaps implemented via a cron script or similar. This can gzip the logs too. Analog has support for these date-based naming schemes, but as per my previous post, doesn't do a complete job of calculating all the names, just the last one. Chris Any views expressed in this message are those of the sender and not necessarily those of CCA Group. The unauthorized use, disclosure, copying or alteration of this message is forbidden. The contents of this message may be confidential and/or privileged, copyright CCA Group and are intended solely for the use of the individual or entity to whom they are addressed. Whilst this message has been scanned, CCA Group cannot guarantee that it is virus free or compatible with your systems and accepts no responsibility for any loss or damage arising from its use. The recipient is advised to run their own anti-virus software. If you receive this message in error please contact [EMAIL PROTECTED] immediately, destroy any copies and delete it from your computer systems. + | TO UNSUBSCRIBE from this list: |http://lists.meer.net/mailman/listinfo/analog-help | | Analog Documentation: http://analog.cx/docs/Readme.html | List archives: http://www.analog.cx/docs/mailing.html#listarchives | Usenet version: news://news.gmane.org/gmane.comp.web.analog.general +
Re: [analog-help] Date based Log File Analysis.
Support for analog web log analyzer wrote: The correct approach would be for analog to loop round generating logfile names from FROM to TO ... It doesn't seem difficult, but I haven't a pressing enough need to put the time in myself - for me it's more of a gotcha than a real problem. If Analog has to loop around generating logfiles names from the From and To, why doesn't it just do the analysis while it's at it? Oh, yeah - it does! Analog analyses log files. If you need logfile management, it's not immediately obvious that it needs to be built in to Analog. Logfile management is necessarily part of the webserver. But having generated daily/weekly/monthly logfiles, how do you efficiently report on an arbitrary period ? Eg - I have monthly archives for some websites, and daily for others. My regular monthly report works fine. To get my regular weekly report I always read the whole month to date, and use a shell script to add in last month too if necessary, letting analog discard the rest of the data. If I want to report occasionally on a larger interval, I have to create a special config file. Currently, the choices seem to be: 1. specify a super-set of the necessary logfiles, and let analog discard what it doesn't need. 2. accept one or more reports based on several subsets. 3. handcraft your config to read the right logfiles. Analog's pretty fast, so maybe (1) is acceptable - takes about .6s per day read. (2) isn't really acceptable. (3) can be done with a perl wrapper, shell script, or something to insert the right LOGFILE statements based on FROM TO, but my perl's not up to the general case. Analog does a great job of allowing from to dates to be specified, so that they don't have to be handcrafted, but something else seems to be required to have it read the right logfiles. Maybe I've missed a trick, but while it's easy to specify the LOGFILE template to analog as a parameterised date, and easy to specify the from and to dates, there's an additional, awkward, step, which analog doesn't do, which is making sure the right logfiles get read for those dates - it just does the TO date. If I wrap analog in something else to do it, I've got to do all the date parameterisation, and analog's own parameterisation doesn't get used. Analog's a great tool, and I like its reports, but I don't have a neat solution to this issue - my script to fiddle the week report seems ugly, and fiddling, say, a quarterly report would be worse. The only place for a neat solution seems to be inside analog, which understands the from to, and the logfile name template, and could, with the addition of a loop, read the exact 7 daily logs required for a weekly report, and 90 days or 3 months for a quarterly report. As I say - maybe I've missed something - if so, please tell me ... Best Regards Chris Any views expressed in this message are those of the sender and not necessarily those of CCA Group. The unauthorized use, disclosure, copying or alteration of this message is forbidden. The contents of this message may be confidential and/or privileged, copyright CCA Group and are intended solely for the use of the individual or entity to whom they are addressed. Whilst this message has been scanned, CCA Group cannot guarantee that it is virus free or compatible with your systems and accepts no responsibility for any loss or damage arising from its use. The recipient is advised to run their own anti-virus software. If you receive this message in error please contact [EMAIL PROTECTED] immediately, destroy any copies and delete it from your computer systems. + | TO UNSUBSCRIBE from this list: |http://lists.meer.net/mailman/listinfo/analog-help | | Analog Documentation: http://analog.cx/docs/Readme.html | List archives: http://www.analog.cx/docs/mailing.html#listarchives | Usenet version: news://news.gmane.org/gmane.comp.web.analog.general +
Re: [analog-help] Date based Log File Analysis.
Support for analog web log analyzer wrote: On Wed, 5 Oct 2005, Chris Sharman wrote: The correct approach would be for analog to loop round generating logfile names from FROM to TO ... It doesn't seem difficult, but I haven't a pressing enough need to put the time in myself - for me it's more of a gotcha than a real problem. Didn't we have this discussion before? Maybe it was with someone else. Possibly. It's not at all easy to do this because even if the logfiles are dated, analog (i) doesn't know the naming scheme used; (ii) doesn't know the rotation schedule, so doesn't know the start date of the logs. You can do it if you have special knowledge about the web server configuration (like, the web server only writes one logfile and logfiles never overlap), but it's almost impossible to do it in general. The naming scheme used is surely known to analog via the LOGFILE statement ? And is surely effectively the same thing as the rotation schedule ? Eg LOGFILE sys$disk:[]access_log_combined_%Y-%M-%D.gz;* would be daily files, from midnight - if they were anything else analog's access to them would already be broken. LOGFILE sys$disk:[]services-%Y-%M.log-gz would be monthly from 00:00 on the 1st. If you generate the logfile name from the template for the from time, and loop round until you get to the to time, you've got all the logs. Your loop step would be determined by which date/time components you found in the LOGFILE statement(s). Overlap could be an issue: you might need an overlap statement, saying how much overlap you cater for. Might be from 0 to a few seconds, I suppose, dependent on your longest request. 0 would probably be good enough for most purposes. Multiple logfiles and the like - would still be allowed as at present by use of OS-dependant wildcards (as in my first example). Regards Chris Any views expressed in this message are those of the sender and not necessarily those of CCA Group. The unauthorized use, disclosure, copying or alteration of this message is forbidden. The contents of this message may be confidential and/or privileged, copyright CCA Group and are intended solely for the use of the individual or entity to whom they are addressed. Whilst this message has been scanned, CCA Group cannot guarantee that it is virus free or compatible with your systems and accepts no responsibility for any loss or damage arising from its use. The recipient is advised to run their own anti-virus software. If you receive this message in error please contact [EMAIL PROTECTED] immediately, destroy any copies and delete it from your computer systems. + | TO UNSUBSCRIBE from this list: |http://lists.meer.net/mailman/listinfo/analog-help | | Analog Documentation: http://analog.cx/docs/Readme.html | List archives: http://www.analog.cx/docs/mailing.html#listarchives | Usenet version: news://news.gmane.org/gmane.comp.web.analog.general +
Re: [analog-help] Analog Help
How do I get rid of following warnings from the erros.txt when i run the analog 6.0? (Note: please see my attached analog.cfg file) C:\dashboard\analog 6.0\analog.exe: Warning C: Unknown configuration command: ignoring it: PAGEALIAS /pkmslogin.form Login There's no pagealias command - perhaps you mean reqalias or some other alias. C:\dashboard\analog 6.0\analog.exe: Warning C: Character 1 not used in REPORTORDER: adding it at end Add 1Qm456 to the end of your reportorder command. | Analog Documentation: http://analog.cx/docs/Readme.html Read the docs ... Chris Any views expressed in this message are those of the sender and not necessarily those of CCA Group. The unauthorized use, disclosure, copying or alteration of this message is forbidden. The contents of this message may be confidential and/or privileged, copyright CCA Group and are intended solely for the use of the individual or entity to whom they are addressed. Whilst this message has been scanned, CCA Group cannot guarantee that it is virus free or compatible with your systems and accepts no responsibility for any loss or damage arising from its use. The recipient is advised to run their own anti-virus software. If you receive this message in error please contact [EMAIL PROTECTED] immediately, destroy any copies and delete it from your computer systems. + | TO UNSUBSCRIBE from this list: |http://lists.meer.net/mailman/listinfo/analog-help | | Analog Documentation: http://analog.cx/docs/Readme.html | List archives: http://www.analog.cx/docs/mailing.html#listarchives | Usenet version: news://news.gmane.org/gmane.comp.web.analog.general +
Re: [analog-help] Re: LogFormat
Support for analog web log analyzer wrote: There are no errors in the log at all - it has some lines stating it turned off some reports, but they are not errors ;) However, it does not find any succesfull requests in the logs - Successful requests: 0 Logfile lines without status code: 36 Corrupt logfile lines: 4 Unwanted logfile entries: 746 FROM -00-00-01 TO -00-00-01 This specifies yesterday, so you've presumably got 746 lines in the file with other dates, which are unwanted. Try DEBUG ON (+V) to see the 4 corrupt lines. Don't know why lines without status code aren't considered corrupt, or how/whether you can see them. I suggest you eyeball them carefully, checking that analog has identified the correct number of lines (for yesterday), and looking for unquoted embedded spaces (which will throw Analog). Correct your webserver config or your analog logformat as appropriate. Chris Any views expressed in this message are those of the sender and not necessarily those of CCA Group. The unauthorized use, disclosure, copying or alteration of this message is forbidden. The contents of this message may be confidential and/or privileged, copyright CCA Group and are intended solely for the use of the individual or entity to whom they are addressed. Whilst this message has been scanned, CCA Group cannot guarantee that it is virus free or compatible with your systems and accepts no responsibility for any loss or damage arising from its use. The recipient is advised to run their own anti-virus software. If you receive this message in error please contact [EMAIL PROTECTED] immediately, destroy any copies and delete it from your computer systems. + | TO UNSUBSCRIBE from this list: |http://lists.meer.net/mailman/listinfo/analog-help | | Analog Documentation: http://analog.cx/docs/Readme.html | List archives: http://www.analog.cx/docs/mailing.html#listarchives | Usenet version: news://news.gmane.org/gmane.comp.web.analog.general +
Re: [analog-help] Question on Analog: Url Parameters
Support for analog web log analyzer wrote: I am trying to create a general report where only the top 50 pages of my site are displayed. I have, unfortunately, encountered two problems. 80% of my site actually is the same Coldfusion page with different url parameters. As an example I might have a page http://domain.com/template.cfm?x=1 and another http://domain.com/template.cfm?x=2 and another http://domain.com/template.cfm?y='test' These should be counted as three different pages. I've used something like filealias /template.cfm?* /template.cfm$$* to achieve what you want - this will make it recognise them all as separate pages. Chris + | TO UNSUBSCRIBE from this list: |http://lists.meer.net/mailman/listinfo/analog-help | | Analog Documentation: http://analog.cx/docs/Readme.html | List archives: http://www.analog.cx/docs/mailing.html#listarchives | Usenet version: news://news.gmane.org/gmane.comp.web.analog.general +
Re: [analog-help] Limiting Time Periods for analysis
,analog-help@lists.meer.net wrote: Hi, I've been asked to report on a user's (mis)activity - we use analog to analyse outgoing web activity - during core work hours - for example, between 09:00 and 12:00 and 14:00 to 17:00, over a period of three months. Because of data volumes (500Mb per day) I've extracted all web access by the user into three monthly files. I could further filter out the log entries from the source files, but hoped that I could do it in Analog. I was hoping that this could be done using the From/To options, or a TIMEINCLUDE but it doesn't look like it. The hourly summary will give you total activity per hour, but if you want the full analog detail (domain, page, etc), I'd recommend using grep or whatever to pre-filter the source files for the hours you want. You don't mention weekends, but they could be a consideration too. Chris + | TO UNSUBSCRIBE from this list: |http://lists.meer.net/mailman/listinfo/analog-help | | Analog Documentation: http://analog.cx/docs/Readme.html | List archives: http://www.analog.cx/docs/mailing.html#listarchives | Usenet version: news://news.gmane.org/gmane.comp.web.analog.general +
Re: [analog-help] Ignore User-Agent
Support for analog web log analyzer wrote: Hi, can i ignore logfile entrys with a specific user-agent (e.g. check_http/1.81+(nagios-plugins+)) ? I'm using analog 6.0. See BROWEXCLUDE in the docs. | Analog Documentation: http://analog.cx/docs/Readme.html Chris + | TO UNSUBSCRIBE from this list: |http://lists.meer.net/mailman/listinfo/analog-help | | Analog Documentation: http://analog.cx/docs/Readme.html | List archives: http://www.analog.cx/docs/mailing.html#listarchives | Usenet version: news://news.gmane.org/gmane.comp.web.analog.general +
Re: [analog-help] Combining pages
Support for analog web log analyzer wrote: Currently the Request Reports for my website are difficult to summarize in Analog, as different addresses for the same page are listed separately and have all to be added together manually. For example I get entries for /default.aspx?Page57.aspxCode=55 and /Page57.aspxCode=55 Can someone tell me some code which will combine such entries, preferably in the shorter form. filealias /default.aspx?* /* All in the docs (see below), under filealias Chris + | TO UNSUBSCRIBE from this list: |http://lists.meer.net/mailman/listinfo/analog-help | | Analog Documentation: http://analog.cx/docs/Readme.html | List archives: http://www.analog.cx/docs/mailing.html#listarchives | Usenet version: news://news.gmane.org/gmane.comp.web.analog.general +
Re: [analog-help] Combining pages
Support for analog web log analyzer wrote: At 08:43 11/07/2005, Chris Sharman wrote about FILEALIAS: You haven't actually quoted my contribution, which was: filealias /default.aspx?* /* My problem seems to be interference between file aliases. I currently have two aliases for my Request Report, namely: FILEALIAS /default.aspx?Page.Aspx* /page.aspx* FILEALIAS page57.Aspx* /page57.aspx* This sorts out most of my problems, but leaves me with a set where I would like to remove the default.aspx? so that these will combine with those listings without the default.aspx? However whenever I try that, it interferes with the above aliases. /default.aspx?Page50.Aspx?Page=50 /default.aspx?Page51.Aspx?page=51 /default.aspx?Page59.Aspx?Genus=Rhododendron /default.aspx?Page57.Aspx?Code=1045 /default.aspx?Page53.Aspx?Group=Devon Can anyone see a way round this? These look like bad urls - the second ? should be or similar. It's not entirely clear exactly what you're trying to achieve, but I'd strongly recommend reading the relevant bits of the very fine docs. Filealias allows multiple wildcards, regular expressions, etc and almost certainly does what you want, but it doesn't work iteratively. The other odd thing is that I have CASE INSENSITIVE in the cfg file but it doesn't seem to work (which probably means I don't understand what it is meant to do) i.e. for example, I am surprised that I have to have the second alias above, but without it the data for page57.Aspx?Code=100 does not combine with page57.aspx?Code=100. Again - read the docs. | Analog Documentation: http://analog.cx/docs/Readme.html If you're going to misquote me, please make sure you misquote me correctly ;-) Chris + | TO UNSUBSCRIBE from this list: |http://lists.meer.net/mailman/listinfo/analog-help | | Analog Documentation: http://analog.cx/docs/Readme.html | List archives: http://www.analog.cx/docs/mailing.html#listarchives | Usenet version: news://news.gmane.org/gmane.comp.web.analog.general +
Re: [analog-help] [analoge help] Reading multiple log files
,analog-help@lists.meer.net wrote: Analog 6.0 - I'm trying to read from about 1000 daily log files. I would like it to read each file and create a report for each day. But I'm having trouble commanding analog to do this automatically. Would appreciate some help with how to insert a line into the CFG so that I wouldn't have to go in and change it for every date's log file. Analog Logfile Documentation: http://analog.cx/docs/logfile.html + | TO UNSUBSCRIBE from this list: |http://lists.meer.net/mailman/listinfo/analog-help | | Analog Documentation: http://analog.cx/docs/Readme.html | List archives: http://www.analog.cx/docs/mailing.html#listarchives | Usenet version: news://news.gmane.org/gmane.comp.web.analog.general + Any views expressed in this message are those of the sender and not necessarily those of CCA Group. The unauthorized use, disclosure, copying or alteration of this message is forbidden. The contents of this message may be confidential and/or privileged, copyright CCA Group and are intended solely for the use of the individual or entity to whom they are addressed. Whilst this message has been scanned, CCA Group cannot guarantee that it is virus free or compatible with your systems and accepts no responsibility for any loss or damage arising from its use. The recipient is advised to run their own anti-virus software. If you receive this message in error please contact [EMAIL PROTECTED] immediately, destroy any copies and delete it from your computer systems. + | TO UNSUBSCRIBE from this list: |http://lists.meer.net/mailman/listinfo/analog-help | | Analog Documentation: http://analog.cx/docs/Readme.html | List archives: http://www.analog.cx/docs/mailing.html#listarchives | Usenet version: news://news.gmane.org/gmane.comp.web.analog.general +
Re: [analog-help] Could someone clarify report differences?
Support for analog web log analyzer wrote: I’m new to analog and I’m hoping someone can clarify the differences between various reports. I’m working at the Adaptive Technology Center at Indiana University, and we are interested in knowing how many successful downloads have been accomplished for a particular .iso file. In both “File Type Report” and “Request Report” the results are the same at 1, 589 requests. However, this is the only file over 100mb on our site, and the “File Size Report” only shows 378 requests. What does this mean? Are any of these results indicating complete, successful downloads? Or do our results show that only 378 of the 1589 requests were successful? Only 378 were 'code 200 success' complete downloads. But there's also '206 partial content' - meaning here's the instalment you asked for. And '304 not modified since last retrieval' meaning yes, your cached copy is still good - you don't need to refresh. Plus assorted failures. If you want a real breakdown of what happened to your 1589 requests, do fileinclude whatever.iso (to exclude everything else), and see what the status code report says for just those. Different servers and clients provide varying support for 206, 304, etc; but for a 100Mb file it's well worth fully supporting both of them. Chris Any views expressed in this message are those of the sender and not necessarily those of CCA Group. The unauthorized use, disclosure, copying or alteration of this message is forbidden. The contents of this message may be confidential and/or privileged, copyright CCA Group and are intended solely for the use of the individual or entity to whom they are addressed. Whilst this message has been scanned, CCA Group cannot guarantee that it is virus free or compatible with your systems and accepts no responsibility for any loss or damage arising from its use. The recipient is advised to run their own anti-virus software. If you receive this message in error please contact [EMAIL PROTECTED] immediately, destroy any copies and delete it from your computer systems. + | TO UNSUBSCRIBE from this list: |http://lists.meer.net/mailman/listinfo/analog-help | | Analog Documentation: http://analog.cx/docs/Readme.html | List archives: http://www.analog.cx/docs/mailing.html#listarchives | Usenet version: news://news.gmane.org/gmane.comp.web.analog.general +
Re: [analog-help] Generating reports for a specific date
,analog-help@lists.meer.net wrote: Hello, I would like to know if I can restrict Analog for a specific date range while generating the reports eg: My webserver access.log(single file) contains logs for 30 days and I want to see the report only for the last 10 days. Is it possible to restrict Analog to generate report for the last 10 days with a single access.log file I am running Analog 4.0 That's a very ancient version (now up to 6.0), but probably FROM yymmdd and TO yymmdd in the config file, or +F and +T on the command line, are what you require. Chris Any views expressed in this message are those of the sender and not necessarily those of CCA Group. The unauthorized use, disclosure, copying or alteration of this message is forbidden. The contents of this message may be confidential and/or privileged, copyright CCA Group and are intended solely for the use of the individual or entity to whom they are addressed. Whilst this message has been scanned, CCA Group cannot guarantee that it is virus free or compatible with your systems and accepts no responsibility for any loss or damage arising from its use. The recipient is advised to run their own anti-virus software. If you receive this message in error please contact [EMAIL PROTECTED] immediately, destroy any copies and delete it from your computer systems. + | TO UNSUBSCRIBE from this list: |http://lists.meer.net/mailman/listinfo/analog-help | | Usenet version: news://news.gmane.org/gmane.comp.web.analog.general | List archives: http://www.analog.cx/docs/mailing.html#listarchives +
Re: [analog-help] Generating reports for a specific date
,analog-help@lists.meer.net wrote: Thanks Chris. I am very sorry, the Analog version I have is 6.0. Can you be more specific about the syntax...I appreciate if you can send more info on this or any link. http://www.analog.cx/docs/include.html#FROMTO The docs are available at http://www.analog.cx/ and also ship with the product, so you may have them locally (which is obviously faster). If so, you can replace http://www.analog.cx/ with your local analog root. Perhaps this url should be included as a FAQ in the footer ? There seem to be quite a lot of questions covered by the docs. Chris On Tue, 21 Jun 2005 Chris Sharman wrote : ,analog-help@lists.meer.net wrote: Hello, I would like to know if I can restrict Analog for a specific date range while generating the reports eg: My webserver access.log(single file) contains logs for 30 days and I want to see the report only for the last 10 days. Is it possible to restrict Analog to generate report for the last 10 days with a single access.log file I am running Analog 4.0 That's a very ancient version (now up to 6.0), but probably FROM yymmdd and TO yymmdd in the config file, or +F and +T on the command line, are what you require. Chris Any views expressed in this message are those of the sender and not necessarily those of CCA Group. The unauthorized use, disclosure, copying or alteration of this message is forbidden. The contents of this message may be confidential and/or privileged, copyright CCA Group and are intended solely for the use of the individual or entity to whom they are addressed. Whilst this message has been scanned, CCA Group cannot guarantee that it is virus free or compatible with your systems and accepts no responsibility for any loss or damage arising from its use. The recipient is advised to run their own anti-virus software. If you receive this message in error please contact [EMAIL PROTECTED] immediately, destroy any copies and delete it from your computer systems. Any views expressed in this message are those of the sender and not necessarily those of CCA Group. The unauthorized use, disclosure, copying or alteration of this message is forbidden. The contents of this message may be confidential and/or privileged, copyright CCA Group and are intended solely for the use of the individual or entity to whom they are addressed. Whilst this message has been scanned, CCA Group cannot guarantee that it is virus free or compatible with your systems and accepts no responsibility for any loss or damage arising from its use. The recipient is advised to run their own anti-virus software. If you receive this message in error please contact [EMAIL PROTECTED] immediately, destroy any copies and delete it from your computer systems. + | TO UNSUBSCRIBE from this list: |http://lists.meer.net/mailman/listinfo/analog-help | | Usenet version: news://news.gmane.org/gmane.comp.web.analog.general | List archives: http://www.analog.cx/docs/mailing.html#listarchives +
[analog-help] browexclude/failrepexclude
I'm seeing errors from a couple of what appear to be cacheing products. I've got pages with a base href=... in to enable me to store all my images on another server. BorderManager 3.0 and HbTools 4.6.2 seem to ignore the base href, and request all these graphics from the wrong website. Both seem to be IE add-ons, and in both cases, the graphics appear to be also requested from the correct website. So, I'd like to throw away all these spurious 404's cluttering my report, without throwing away any of the successful page requests, and just for these browser strings. BROWEXCLUDE would throw away the successes too, and FAILREPEXCLUDE won't let me act on the user agent string. Is something clever with regular expressions possible ? Obviously I'd like to know of any 'real' 404's from other browsers. I guess I need a regular expression which works on the whole line, not just particular fields. Thanks Chris Any views expressed in this message are those of the sender and not necessarily those of CCA Group. The unauthorized use, disclosure, copying or alteration of this message is forbidden. The contents of this message may be confidential and/or privileged, copyright CCA Group and are intended solely for the use of the individual or entity to whom they are addressed. Whilst this message has been scanned, CCA Group cannot guarantee that it is virus free or compatible with your systems and accepts no responsibility for any loss or damage arising from its use. The recipient is advised to run their own anti-virus software. If you receive this message in error please contact [EMAIL PROTECTED] immediately, destroy any copies and delete it from your computer systems. + | TO UNSUBSCRIBE from this list: |http://lists.meer.net/mailman/listinfo/analog-help | | Usenet version: news://news.gmane.org/gmane.comp.web.analog.general | List archives: http://www.analog.cx/docs/mailing.html#listarchives +
Re: AW: [analog-help] Logfile format
FROM 050201 TO 050401 But the report shows only: Analyzed requests from Thu, Mar 31 2005 at 5:21 AM to Fri, Apr 01 2005 at 11:58 PM (1.78 days). If you've got your logfile broken into days, or similar, and have used LOGFILE blah-%Y-%m-%d or something like that, then analog will only read the logfile pointed at by the TO date - it won't loop through other dates picking up other log files. I use LOGFILE blah-%Y-%m-* to pick up whole months - you could doubtless do something similar to have it read the whole year, and report on Feb/Mar. I made an enhancement suggestion/request here the other day on this subject. Chris Any views expressed in this message are those of the sender and not necessarily those of CCA Group. The unauthorized use, disclosure, copying or alteration of this message is forbidden. The contents of this message may be confidential and/or privileged, copyright CCA Group and are intended solely for the use of the individual or entity to whom they are addressed. Whilst this message has been scanned, CCA Group cannot guarantee that it is virus free or compatible with your systems and accepts no responsibility for any loss or damage arising from its use. The recipient is advised to run their own anti-virus software. If you receive this message in error please contact [EMAIL PROTECTED] immediately, destroy any copies and delete it from your computer systems. + | TO UNSUBSCRIBE from this list: |http://lists.meer.net/mailman/listinfo/analog-help | | Usenet version: news://news.gmane.org/gmane.comp.web.analog.general | List archives: http://www.analog.cx/docs/mailing.html#listarchives +
Re: [analog-help] Zip log files and Analog on VMS
Support for analog web log analyzer wrote: Anyone been able to have analog read zip files on VMS? Reads gzipped files - I've never tried plain zip. Chris Any views expressed in this message are those of the sender and not necessarily those of CCA Group. The unauthorized use, disclosure, copying or alteration of this message is forbidden. The contents of this message may be confidential and/or privileged, copyright CCA Group and are intended solely for the use of the individual or entity to whom they are addressed. Whilst this message has been scanned, CCA Group cannot guarantee that it is virus free or compatible with your systems and accepts no responsibility for any loss or damage arising from its use. The recipient is advised to run their own anti-virus software. If you receive this message in error please contact [EMAIL PROTECTED] immediately, destroy any copies and delete it from your computer systems. + | TO UNSUBSCRIBE from this list: |http://lists.meer.net/mailman/listinfo/analog-help | | Usenet version: news://news.gmane.org/gmane.comp.web.analog.general | List archives: http://www.analog.cx/docs/mailing.html#listarchives +
Re: [MAYBE SPAM] Re: [analog-help] Excluding log records by 'bytes transferred'
,analog-help@lists.meer.net wrote: I don't know how to exclude records entirely based on bytes, but you can use a FLOOR command to control what is shown within certain reports. For example, to control what is shown in the Domain Report, you could use DOMFLOOR 141b to list in the report only files with at least 141 bytes transferred. The Analog docs discuss FLOOR commands in three places: * http://www.analog.cx/docs/othreps.html#FLOOR * http://www.analog.cx/docs/hierreps.html#SUBFLOOR * http://www.analog.cx/docs/hierreps.html#ARGSFLOOR My understanding was that that would list domains with a total of at least 141 bytes - and that 10 * 20 byte requests would qualify ?? Chris Any views expressed in this message are those of the sender and not necessarily those of CCA Group. The unauthorized use, disclosure, copying or alteration of this message is forbidden. The contents of this message may be confidential and/or privileged, copyright CCA Group and are intended solely for the use of the individual or entity to whom they are addressed. Whilst this message has been scanned, CCA Group cannot guarantee that it is virus free or compatible with your systems and accepts no responsibility for any loss or damage arising from its use. The recipient is advised to run their own anti-virus software. If you receive this message in error please contact [EMAIL PROTECTED] immediately, destroy any copies and delete it from your computer systems. + | TO UNSUBSCRIBE from this list: |http://lists.meer.net/mailman/listinfo/analog-help | | Usenet version: news://news.gmane.org/gmane.comp.web.analog.general | List archives: http://www.analog.cx/docs/mailing.html#listarchives +
Re: [MAYBE SPAM] [analog-help] Status Code Report
,analog-help@lists.meer.net wrote: Hi all, Quick query - I can generate the State Code report for one of our websites quite easily (just set 'STATUS ON' in analog.cfg)... however, our users want a more detailed report whereby they can examine, for example, all 404 Error Requests - ie. find out exactly what documents couldn't be found, rather than just a numbered summary. Is this possible with Analog? The failure report, together with STATUSINCLUDE/EXCLUDE if you need to drop other failures. By the way, does anyone know why some messages arrive from ,analog-help@ rather than just analog-help@ ? Means I can't reply directly. Have I got something misconfigured in my mail client (moz/w2k) ? Chris Any views expressed in this message are those of the sender and not necessarily those of CCA Group. The unauthorized use, disclosure, copying or alteration of this message is forbidden. The contents of this message may be confidential and/or privileged, copyright CCA Group and are intended solely for the use of the individual or entity to whom they are addressed. Whilst this message has been scanned, CCA Group cannot guarantee that it is virus free or compatible with your systems and accepts no responsibility for any loss or damage arising from its use. The recipient is advised to run their own anti-virus software. If you receive this message in error please contact [EMAIL PROTECTED] immediately, destroy any copies and delete it from your computer systems. + | TO UNSUBSCRIBE from this list: |http://lists.meer.net/mailman/listinfo/analog-help | | Usenet version: news://news.gmane.org/gmane.comp.web.analog.general | List archives: http://www.analog.cx/docs/mailing.html#listarchives +
Re: [MAYBE SPAM] [analog-help] Log file problems - possible double counting
Support for analog web log analyzer wrote: Hello I am running reports on some log files I've been given. They seem sequential and from the same time period but I keep getting the following errors: /usr/local/analog/analog-6.0/analog: Warning D: In Browser Report, SORTBY (pages) doesn't match FLOOR (requests) /usr/local/analog/analog-6.0/analog: Warning L: logfiles /logs/webserver/server1/de/http/access.complete.20050201.Z and /logs/webserver/server1/de/http/access.complete.20050202.Z overlap: possible double counting It seems to be suggesting that the logs for subsequent days share at least some of the same information. Does anyone know of a method to override these errors and get the reports to run? The reports do run - it's only a warning, so you should decide whether you're happy with the resulting stats, or not - but they should be there. You should fix the first warning - browser sortby/floor discrepancy. It would be nice if analog (or some other tool) could do a more careful check on 'possible duplicates' - I see the same (from our demon.co.uk hosted site) - could/has that been done ? It shouldn't be too onerous to check any 'out of period' lines for exact duplication, should it ? Chris --- Any views expressed in this message are those of the sender and not necessarily those of CCA Group. The unauthorized use, disclosure, copying or alteration of this message is forbidden. The contents of this message may be confidential and/or privileged, copyright CCA Group and are intended solely for the use of the individual or entity to whom they are addressed. Whilst this message has been scanned, CCA Group cannot guarantee that it is virus free or compatible with your systems and accepts no responsibility for any loss or damage arising from its use. The recipient is advised to run their own anti-virus software. If you receive this message in error please contact [EMAIL PROTECTED] immediately, destroy any copies and delete it from your computer systems. + | TO UNSUBSCRIBE from this list: |http://lists.meer.net/mailman/listinfo/analog-help | | Usenet version: news://news.gmane.org/gmane.comp.web.analog.general | List archives: http://www.analog.cx/docs/mailing.html#listarchives +
Re: [MAYBE SPAM] [analog-help] creating unique image names
Support for analog web log analyzer wrote: I want to create unique file names for the charts created by analog, such as isearchq.png. I am unable to locate the configuration parameter to do this. chartdir and localchartdir (url and matching writable folder) Chris --- Any views expressed in this message are those of the sender and not necessarily those of CCA Group. The unauthorized use, disclosure, copying or alteration of this message is forbidden. The contents of this message may be confidential and/or privileged, copyright CCA Group and are intended solely for the use of the individual or entity to whom they are addressed. Whilst this message has been scanned, CCA Group cannot guarantee that it is virus free or compatible with your systems and accepts no responsibility for any loss or damage arising from its use. The recipient is advised to run their own anti-virus software. If you receive this message in error please contact [EMAIL PROTECTED] immediately, destroy any copies and delete it from your computer systems. + | TO UNSUBSCRIBE from this list: |http://lists.meer.net/mailman/listinfo/analog-help | | Usenet version: news://news.gmane.org/gmane.comp.web.analog.general | List archives: http://www.analog.cx/docs/mailing.html#listarchives +
Re: [analog-help] Problem with Form and CGI on TOmcat 4.1 with Analog 6.0
,analog-help@lists.meer.net wrote: 1. Is there any Java implementation for using the Analog? We have a web based admin system. I need to provide web based GUI in which the Admin user selects all the commands like DAILYREP etc., and ask for a report. The server should have some Servlet kind of program, which can contact the Analog, generate a report and display it to the user. I could achive this with the HTML form and Perl program that are supplied as samples along with the install kit of Analog. But I need one with Java. I am looking into that Perl program, but its greek and latin to me, as I never knwo abt Perl. No idea ( don't know java) - but I think you'll find helpful examples for assorted environments either in the full analog kit or on the analog website. 2. Need another suggestion in General Summary. I file based report like Number of HTML requests: 1042035 Number of script requests: 69784 Number of non-HTML requests: 0 Number of malformed requests (all dates): 494865 Sounds more like the file type report section: reqs%reqs Mbytes %bytes extension 64630 49.76% 222.34 40.11% .jpg [JPEG graphics] 62654.82% 106.69 19.25% [directories] 45909 35.35% 100.74 18.18% .gif [GIF graphics] 61674.75% 82.63 14.91% .js [JavaScript code] 42213.25% 34.42 6.21% [no extension] 23181.78% 4.350.78% .css [Cascading Style Sheets] 32 0.02% 2.440.44% .pdf [Adobe Portable Document Format] 198 0.15% 0.640.11% .html [Hypertext Markup Language] 135 0.10% 0.030.01% [not listed: 2 extensions] Analog is a complex product, with a lot of config options - you really do need to read the documentation. Chris --- Any views expressed in this message are those of the sender and not necessarily those of CCA Group. The unauthorized use, disclosure, copying or alteration of this message is forbidden. The contents of this message may be confidential and/or privileged, copyright CCA Group and are intended solely for the use of the individual or entity to whom they are addressed. Whilst this message has been scanned, CCA Group cannot guarantee that it is virus free or compatible with your systems and accepts no responsibility for any loss or damage arising from its use. The recipient is advised to run their own anti-virus software. If you receive this message in error please contact [EMAIL PROTECTED] immediately, destroy any copies and delete it from your computer systems. + | TO UNSUBSCRIBE from this list: |http://lists.meer.net/mailman/listinfo/analog-help | | Usenet version: news://news.gmane.org/gmane.comp.web.analog.general | List archives: http://www.analog.cx/docs/mailing.html#listarchives +
Re: [analog-help] Problem with Form and CGI on TOmcat 4.1 with Analog 6.0
,analog-help@lists.meer.net wrote: Hi, I am new to Analog. I was trying out the Analog5.0.1. I ran that program on the local machine and also deployed the form and perl program on to Tomcat 4.1. This ran successfully and I gave a demo to my boss also. Then we decided to go for the latest verion Analog6.0. I have setup the Analog 6.0 on the local machine and successfully generated the report. Now I am trying to deploy the same on the Tomcat 4.1. I have made the necessary changes to the HTML and Perl program. When I submit the form, a blank HTML page is displayed and the report HTML source code is written to the TOmcat Log. I am wondering what could be the problem. Please help me in this rehgard. Don't know what 'form and perl' you're talking about, but the analog output was extensively enhanced (to xhtml+css by default) between 5 6. XML and HTML are available options (use the OUTPUT command). Your perl probably needs changing to process it correctly. If that's not a option, you could try OUTPUT HTML to go back to html output (though I don't know if it's compatible with 5.01), or simply stick with 5.01. Chris --- Any views expressed in this message are those of the sender and not necessarily those of CCA Group. The unauthorized use, disclosure, copying or alteration of this message is forbidden. The contents of this message may be confidential and/or privileged, copyright CCA Group and are intended solely for the use of the individual or entity to whom they are addressed. Whilst this message has been scanned, CCA Group cannot guarantee that it is virus free or compatible with your systems and accepts no responsibility for any loss or damage arising from its use. The recipient is advised to run their own anti-virus software. If you receive this message in error please contact [EMAIL PROTECTED] immediately, destroy any copies and delete it from your computer systems. + | TO UNSUBSCRIBE from this list: |http://lists.meer.net/mailman/listinfo/analog-help | | Usenet version: news://news.gmane.org/gmane.comp.web.analog.general | List archives: http://www.analog.cx/docs/mailing.html#listarchives +
Re: [analog-help] missing images
Support for analog web log analyzer wrote: I`m using apache server, I`m running the log file fine but I have missing images how do I config where the images are? I found this in the help IMAGEDIR img/ # relative URL: within the same directory as the output IMAGEDIR /img/ # off the root directory of your server IMAGEDIR http://www.myother.server.com/img/ # on another server but I`m not sure where to add this information to i add it to the .cfg file? Also where are the images anyway, do I have to configure something else? These are the bar* images (for drawing bar graphs, included in complete analog distributions) and the analogo image. Make sure they're somewhere they can be referenced with a url, decide where you're putting the analog output [x]html, and set imagedir appropriately. The default seems to be images/ - eg images/barb1.png If you don't have it set up correctly, I guess you'll see '+' characters for your bar graphs, and the other logos will be missing. Yes, it goes in a .cfg file. chartdir localchartdir control pie charts, if those are the images you're missing. Chartdir is like imagedir; localchartdir gives the start of a filespec on your system, which corresponds to the url generated with chartdir. Analog will create pie chart graphics here (must be writable). It's up to you to specify chartdir localchartdir to avoid any naming conflicts between different analog reports. Chris --- Any views expressed in this message are those of the sender and not necessarily those of CCA Group. The unauthorized use, disclosure, copying or alteration of this message is forbidden. The contents of this message may be confidential and/or privileged, copyright CCA Group and are intended solely for the use of the individual or entity to whom they are addressed. Whilst this message has been scanned, CCA Group cannot guarantee that it is virus free or compatible with your systems and accepts no responsibility for any loss or damage arising from its use. The recipient is advised to run their own anti-virus software. If you receive this message in error please contact [EMAIL PROTECTED] immediately, destroy any copies and delete it from your computer systems. + | TO UNSUBSCRIBE from this list: |http://lists.meer.net/mailman/listinfo/analog-help | | Usenet version: news://news.gmane.org/gmane.comp.web.analog.general | List archives: http://www.analog.cx/docs/mailing.html#listarchives +
[analog-help] Long lines in Anlog 6.0
I've just upgraded from 5.32 to 6.0. My immediate stats page has stopped working. This is due to an incompatibility between Analog 6.0 my webserver: OSU HTTP_SERVER on VMS. It works fine serving html files with very long lines; it only breaks when piping output directly from analog to the browser. The gotos line generated by analog is now 1150 characters long (it used to be 800-odd). While this isn't an error, it falls foul of the 1024 character limit imposed by my environment (which I can't change). It would be a very helpful improvement to break such very long lines (255 would probably be a good upper bound to work to for readability reasons). I notice that the leading colon gets translated via the language file, but unfortunately the pipe separator doesn't - if it did I might be able to fix this in the language file (if it were possible to add a real line break - which I haven't checked). I'll add my own line break next to the pipe in the source. I'd be very grateful if a similar enhancement could make it to the next edition. Thanks, Chris Sharman --- Any views expressed in this message are those of the sender and not necessarily those of CCA Group. The unauthorized use, disclosure, copying or alteration of this message is forbidden. The contents of this message may be confidential and/or privileged, copyright CCA Group and are intended solely for the use of the individual or entity to whom they are addressed. Whilst this message has been scanned, CCA Group cannot guarantee that it is virus free or compatible with your systems and accepts no responsibility for any loss or damage arising from its use. The recipient is advised to run their own anti-virus software. If you receive this message in error please contact [EMAIL PROTECTED] immediately, destroy any copies and delete it from your computer systems. + | TO UNSUBSCRIBE from this list: |http://lists.meer.net/mailman/listinfo/analog-help | | Usenet version: news://news.gmane.org/gmane.comp.web.analog.general | List archives: http://www.analog.cx/docs/mailing.html#listarchives +
Re: [analog-help] Long lines in Anlog 6.0
Support for analog web log analyzer wrote: This is due to an incompatibility between Analog 6.0 my webserver: OSU Oh no it isn't, it's because my config language files had the wrong protection (again). Surprisingly, the webserver copes with the long lines - although I'm sure I've seen problems with that in the past. Breaking up the gotos into more manageable linelengths would be tidy though. The next longest are the 'bar' lines, at 300-odd - much more reasonable. I suspect it would be a bad idea to introduce whitespace between the images there - although it would probably be ok inside the image tag. Sorry, Chris --- Any views expressed in this message are those of the sender and not necessarily those of CCA Group. The unauthorized use, disclosure, copying or alteration of this message is forbidden. The contents of this message may be confidential and/or privileged, copyright CCA Group and are intended solely for the use of the individual or entity to whom they are addressed. Whilst this message has been scanned, CCA Group cannot guarantee that it is virus free or compatible with your systems and accepts no responsibility for any loss or damage arising from its use. The recipient is advised to run their own anti-virus software. If you receive this message in error please contact [EMAIL PROTECTED] immediately, destroy any copies and delete it from your computer systems. + | TO UNSUBSCRIBE from this list: |http://lists.meer.net/mailman/listinfo/analog-help | | Usenet version: news://news.gmane.org/gmane.comp.web.analog.general | List archives: http://www.analog.cx/docs/mailing.html#listarchives +
