Browsera Help

General

Test Report

General

How is Browsera different from Browsershots, Adobe BrowserLab, or Microsoft SuperPreview?

Browsera is meant to be used as a testing tool, not just a screenshot tool, and works at the site-level instead of the page level. Browsershots, BrowserLab, and SuperPreview provide multiple screenshots of a single page, and leave the rest up to you. Browsera will crawl the pages on your site and actually test them for layout inconsistencies and scripting errors. Instead of spending time identifying all of your URLs and reviewing each screenshot for each page, you can simply view a consolidated Browsera report listing the specific pages and browsers that have problems.

What browsers can I test?

We are currently supporting the following browsers:
  • Chrome 53.0 / Windows
  • Firefox 44.0 / Windows
  • Firefox 45.0 / Windows
  • Firefox 46.0 / Windows
  • Firefox 47.0 / Windows
  • Internet Explorer 10.0 / Windows
  • Internet Explorer 11.0 / Windows
  • Internet Explorer 6.0.3790.3959 / Windows
  • Internet Explorer 7.0.5730.13 / Windows
  • Internet Explorer 8.0.6001.18702 / Windows
  • Internet Explorer 9.0.8112.16421 / Windows
  • Safari 10.0 / Mac
  • Safari 6.0 / Mac
  • Safari 7.0 / Mac
  • Safari 8.0 / Mac
  • Safari 9.0 / Mac

How well does it work?

Our technology continues to evolve, but we're already pretty good at finding the most common classes of problems caused by browser inconsistencies. Every web design implementation has a different 'signature' though, so we cannot guarantee it will work as well for everyone. We can say that tests against large websites have shown very accurate and promising results in locating bugs.

How do I submit my site for testing?

Sign up and begin testing right away.

What IP addresses does your service use?

We currently use these IPs:
  • 198.58.97.30
  • 66.114.148.226 - 66.114.148.230
However, during periods of peak demand we may use additional machines from the cloud with different IP addresses.

When will you add browser XYZ?

If you're interested in having us support another browser, just contact us and let us know what you need. We'll let you know what our plans are for that browser.

Do you support FRAMES and IFRAMES?

Web pages containing FRAMES and IFRAMES have limited support. Due to browser security constraints, we are unable to examine the DOM contents inside of frames, and thus only compare layouts of the outer page.

Do you support Flash/Air/SilverLight?

We do not currently support these browser plugin technologies due to the occasional instability they cause in our browser environment. We may add support for them in the future.

How do I test internal or private sites?

Browsera can only test sites that are internet facing. However, we support HTTP authentication, and you can deploy your application requiring a login. For additional security, you may restrict access to our IPs, which are currently:
  • 198.58.97.30
  • 66.114.148.226 - 66.114.148.230
However, during periods of peak demand we may use additional machines from the cloud with different IP addresses.

Does Browsera have an API?

Yes, Browsera has an API which allows you to manage and create tests via a REST XML API. See API Documentation for more information.

Test Report

Why are my pages or browsers timing out?

Some of our browsers may time out when run against your site. We load all browsers simultaneously to ensure each page is as consistent across browsers. If the "Baseline" browser you selected timed out, the other browsers are cancelled because we do not have a baseline to compare to. This is a page timeout. Othwerwise, if other, non-baseline, browsers timeout, we continue testing the page. Here are the most common reasons why these timeouts occur:
  • The entire page contents are taking more than 45 seconds for us to download and render.
  • The page loaded, but was then redirected on the client side using Javascript or a META refresh.
  • The SSL security certificate was not valid.
  • Scripting on the page has not completed in the 45 seconds.
  • The page or one of its resources could not be reached. (All content must load successfully.)
  • An error occurred on our end.
If a page or browser continues to time out for you, please contact us and we will investigate the issue.

Why are there false positives?

Sometimes Browsera says there is a difference, but there isn't. The elements it claims are different actually look the same. This happens because some browsers report positions differently than others. For example, one browser may report a DIV element as filling the entire page, while another reports it as a smaller box. You can often see this for yourself using tools like Firebug or IE Developer Toolbar by inspecting the element in question. We're working on improving the number of these false positives, but in the meantime, you can click on the 'X' next to the difference to delete and exclude it from the current results, as well as any future results.

Why aren't all elements being matched to the baseline browser?

We match corresponding elements between browsers in order to compare their layouts. Our algorithm uses the structure of the DOM to correlate elements, thus, if you have different HTML structure on the page, we aren't able to determine which elements match with which. In most cases this is a result of either improperly closed tags or illegally nested tags.

Sometimes, the contents are just different. For example, if a div has the structure <div><img><img></div> in the baseline browser, but <div><img></div> in another, we are unable to correlate the second <img> tag. In cases where fewer than 85% of the elements match, we flag the browser for manual review.

Also, browser-specific content (such as an unsupported browser message) can cause this behavior.

To improve your match percentage and Browsera's effectiveness, we recommend keeping your structural HTML as similar as possible for each browser and ensure all tags are closed and nested properly.

What is a Unique Layout Difference?

Unique Layout Differences are the total number of layout differences detected across the entire test. Differencess appearing on multiple pages are counted just once. For example, if you had a common header on your site and five pages had the same difference detected in the header, this would count as only one unique difference.

What Layout Differences do you detect?

Our testing algorithms analyze the page looking for discrepancies the way a human would. For example, we look for items that have excess line breaks, or elements that appear to be shifted relative to others on the page. We ignore differences that are expected, such as slightly different font sizes. We're constantly refining our algorithms and would love to hear feedback on any false positives or issues we've missed. Here are a few things we aren't currently checking:
  • Alpha transparency (particularly on IE6)
  • Element z-indexing
  • Hidden or off-screen contents