Hi everyone, new here and to using Imperva. I've been working with it for a few weeks now for a client.
I find the web interface, especially the policy management and event logs, to be really slow to load. Sometimes there's a spinner for 10-15 seconds just to filter logs. Is this a common experience, or could it be something on my setup? Coming from other SaaS tools, it feels noticeably laggy.
Just trying to figure out if it's me or the platform. Thanks for any insights!
That's interesting you mention this, because I've been seeing something similar, but not exactly the same. I've been working primarily with the reporting and dashboard sections, and I've noticed the slowness seems to depend heavily on the time range selected. If I'm pulling data for just the last hour, it's reasonably quick. But if I need to look back over a 24-hour period or filter by a specific blocked attack type, the interface can become very unresponsive, much like you're describing.
I had initially assumed it was due to the volume of data being processed for my particular client's traffic, but your post makes me wonder if it's more widespread. Have you noticed any pattern with the amount of log data your client generates correlating with the lag, or is it slow even for very simple, immediate-timeframe tasks?
You're definitely not alone in noticing that lag. I've done some systematic testing on the event log page because the 10-15 second spinner was impacting our team's workflow. The delay isn't just about your setup, it's largely tied to how the UI queries the backend.
I found the primary bottleneck is the initial aggregation call that runs before the table data is fetched. When you apply a filter, the UI first requests a count of matching records across all pages. That call seems to run a full scan on the log database for your selected time range. If you're looking at a busy site over 24 hours, that's a heavy operation. Other SaaS tools often cheat by returning an estimated count or paginating without the upfront total scan.
Try this: open your browser's developer tools, go to the Network tab, and filter for XHR requests. You'll see a `count` or `aggregate` call that takes most of those seconds, followed by the actual `data` fetch. The workaround is to use the predefined quick filters (like "Last 1 hour") when possible, as they might use cached aggregates.