Use Powershell v3. http://jdhitsolutions.com/blog/2012/12/friday-fun-scraping-the-web-with-powershell-v3/
>From https://www.google.com/search?btnG=1&pws=0&q=invoke-webrequest+scrape (Note the PoSH 3 is not installed on Win7 by default.) On Fri, Nov 1, 2013 at 11:22 AM, Jimmy Tran <[email protected]> wrote: > Ok, so lets ignore number 3. How am I supposed to search the strings? > Am I running a script or an app? Can you give me some insight on what you > are suggesting so I can research this? I only have a few hours left to get > this accomplished.**** > > ** ** > > Thanks,**** > > > jimmy**** > > **** > > ** ** > > *From:* [email protected] [mailto: > [email protected]] *On Behalf Of *Daniel Chenault > *Sent:* Friday, November 01, 2013 7:55 AM > *To:* [email protected] > > *Subject:* Re: [NTSysADM] RE: test IE uptime?**** > > ** ** > > 3. The list of running apps does not show status nor is that a very > reliable way of determining if an app is truly hung. Only thing I can think > of here is a blind restart of IE on a periodic basis. Perhaps the IE SDK > has a clue (don’t have this myself)**** > > **** > > 4. Then that’s easy; only two strings to search for. “511 Transit > Departures” = success, “sign problem” indicates a known failure state while > anything else is an unknown failure state. **** > > **** > > *From:* Jimmy Tran <[email protected]> **** > > *Sent:* Friday, November 01, 2013 9:03 AM**** > > *To:* [email protected] **** > > *Subject:* RE: [NTSysADM] RE: test IE uptime?**** > > **** > > 3. Does a list of running apps still apply is IE is > frozen/stuck/non-responsive. I guess that would need to be tested but I > don’t know how to create such a failure.**** > > **** > > 4. If the page were to load, it would say “511 Transit Departures” in the > header. If it doesn’t load, it would show the default page not found be IE > when there is no network connectivity. If the 511.org site is down, or > the specific page for that station is down, it would say “Sign Problem. Try > calling 511 and say “Departure Times”**** > > **** > > I don’t have any control over the page, and I don’t know what language was > used to build the page. As unfortunately, I don’t think the stakeholders > know the answer to that either.**** > > **** > > *From:* [email protected] [ > mailto:[email protected] <[email protected]>] *On > Behalf Of *Daniel Chenault > > *Sent:* Friday, November 01, 2013 6:36 AM > *To:* [email protected] > > *Subject:* Re: [NTSysADM] RE: test IE uptime?**** > > **** > > Okay, MUCH better. Since the need has shown itself to be mission-critical > perhaps an app would be better than a script; provides more functionality > to the developer.**** > > **** > > 1. As you say, from the system log **** > > 2. Ping the host. Fail is written to a local log for later transmittal.*** > * > > 3. Check running list of apps**** > > 4. Is there a word or group of words in the webpage that ALWAYS appears on > successful load but NEVER appears in a failed load? More to the point, are > you in control of the page’s contents? What language is used to build the > page – perhaps the last command in the load could set a flag; then you can > check for that.**** > > **** > > *From:* Jimmy Tran <[email protected]> **** > > *Sent:* Thursday, October 31, 2013 11:39 PM**** > > *To:* [email protected] **** > > *Subject:* RE: [NTSysADM] RE: test IE uptime?**** > > **** > > So let me explain the situation.**** > > **** > > These are 40” displays connected to a w7 box all enclosed inside of a > protective case. The displays are at train stations, bus stops, and light > rail stations. If the machine were to be turned on, it would power up, > wait for network connectivity and start a scheduled task which would load > IE in kiosk mode. IE would have a specific home page set for that > location. Then the page refreshes every 10 seconds to update the display. > **** > > **** > > We have no control or access to the 511.org server who hosts the webpages > and transit times. All we have access to is the standalone kiosk. The > test is to determin iif the home page ever fails to refresh/reload. We > just want to make sure if the page does not load for whatever reason it > might be, that it is logged and documented.**** > > **** > > If it were to fail, it would be due to the following reasons:**** > > 1. Machine loss power – system up time would tell us that**** > > 2. 4g or wifi connection went out – currently no way to log this**** > > 3. IE process crashed – currently no way to log this**** > > 4. 511 page is offline – currently no way to log this.**** > > **** > > So to sum it up, I need to find a solution that can answer items 2,3, and > 4. I figured a script running on the host that could somehow search the > page for key words and if it fails, it records it to a txt file.**** > > **** > > Jimmy**** > > **** > > *From:* [email protected] [ > mailto:[email protected] <[email protected]>] *On > Behalf Of *Richard Stovall > > *Sent:* Thursday, October 31, 2013 7:14 PM > *To:* [email protected] > > *Subject:* Re: [NTSysADM] RE: test IE uptime?**** > > **** > > What type of error?**** > > **** > > There's almost certainly a way to do what you're looking for with wget, or > Powershell, or another scripting language, if what you're really trying to > gauge is what the server is doing. I think that's the key question at the > moment. Are you trying to record and interpret a particular client > response, or are you actually looking to make sure the server is doing what > *it* is supposed to be doing. (Or some combination of both, which might > entail separate monitoring straegies.)**** > > **** > > On Thu, Oct 31, 2013 at 10:04 PM, Jimmy Tran <[email protected]> > wrote:**** > > That was just a political requirement made by people who don't understand > computers. I'm not sure if other browsers can run in kiosk mode. > > The main thing we need to know is if the webpage displays an error during > a period of 7 days we need to be notified. > > What were you thinking? **** > > > Sent from my Samsung Galaxy S4 **** > > -----Original Message----- > *From:* Richard Stovall [[email protected]]**** > > *Received:* Thursday, 31 Oct 2013, 6:57pm > *To:* [email protected] [[email protected]] > *Subject:* Re: [NTSysADM] RE: test IE uptime?**** > > Let me rephrase the question.**** > > **** > > Does IE *have* to be the rendering engine?**** > > **** > > If yes, why? Are you testing IE's (a client's) ability to render a server > response correctly, or are you really trying to monitor the server itself? > There's likely a big difference here in terms of what is easy to do and > what is not. **** > > **** > > On Thu, Oct 31, 2013 at 9:49 PM, Jimmy Tran <[email protected]> wrote: > **** > > Yes **** > > > > Sent from my Samsung Galaxy S4 **** > > -----Original Message----- > *From:* Richard Stovall [[email protected]] > *Received:* Thursday, 31 Oct 2013, 6:45pm > *To:* [email protected] [[email protected]]**** > > *Subject:* Re: [NTSysADM] RE: test IE uptime?**** > > Does IE have to be the rendering engine?**** > > **** > > On Thu, Oct 31, 2013 at 9:05 PM, Jimmy Tran <[email protected]> wrote: > **** > > If there's a cost associated, that may be ok. There really isn't a budget > allocated for this but if it's cheaper to buy a software than me writing > some scripts that might be a better way to go. > > You provide some great keywords to do some searching on thanks! **** > > > > Sent from my Samsung Galaxy S4 **** > > -----Original Message----- > *From:* Ken Schaefer [[email protected]] > *Received:* Thursday, 31 Oct 2013, 5:56pm > *To:* [email protected] [[email protected]] > *Subject:* RE: [NTSysADM] RE: test IE uptime?**** > > Well, the basic concept you are looking at is “end user experience > monitoring”, as opposed to “component monitoring” (e.g. a server or hard > disk) or “service monitoring” (e.g. is a website up)**** > > **** > > There’s plenty of third party products out there that can simulate client > actions – usually via an agent installed on the remote machine. Or you do a > “poor man’s” job and do this yourself with a bit of scripting. I assumed > you wanted the latter since you didn’t mention any existing tools you have > or any budget.**** > > **** > > Cheers**** > > Ken**** > > **** > > *From:* [email protected] [mailto: > [email protected]] *On Behalf Of *Jimmy Tran > *Sent:* Friday, 1 November 2013 11:45 AM > *To:* [email protected] > *Subject:* RE: [NTSysADM] RE: test IE uptime?**** > > **** > > I will look in to that. Any other ideas? **** > > > Sent from my Samsung Galaxy S4 > > -----Original Message----- **** > > *From:* Ken Schaefer [[email protected]] > *Received:* Thursday, 31 Oct 2013, 5:30pm > *To:* [email protected] [[email protected]] > *Subject:* [NTSysADM] RE: test IE uptime? **** > > Write a little script (e.g. using VBScript and ServerXMLHTTP object) that > makes the same HTTP request? You can then log the details to a log file to > later analyse.**** > > **** > > Cheers**** > > Ken**** > > **** > > *From:* [email protected] [ > mailto:[email protected] <[email protected]>] *On > Behalf Of *Jimmy Tran > *Sent:* Friday, 1 November 2013 11:21 AM > *To:* [email protected] > *Subject:* [NTSysADM] test IE uptime?**** > > **** > > I’ve been tasked with determining a way to document if a web page in IE > errors out. Basically, the page will refresh every 10 seconds with updated > data. I need to find a way to see if the page comes back with a page error > or not. Any ideas out there? These tests will be performed at kiosks that > have no management that have poor internet connectivity.**** > > **** > > All I can currently measure is system up time, not internet connectivity > uptime or page errors in IE.**** > > **** > > TIA**** > > **** > > Jimmy **** > > **** > > **** > > **** > > **** >
