Table of Contents
Episode Summary
In this episode, our host Bret Alarcon is joined by Senior Support Specialist, Nick Salvatori, and Tech. Ops Consultant, Ross Tenaglia to discuss server maintenance for on-premise customers. The group touches on the importance of staying up-to-date, creating frequent backups, and managing server disk space.
Recording
Transcript
Ross Tenaglia 0:09
Hello, and welcome to this week's edition of RecChat. I'm your host Bret Alarcon. So Ross is here today. And he's going to be talking about tuning up your RecTrac and WebTrac servers. So we're going to be covering how to recover some disk space, some server optimization tips, managing your database app servers, and then we'll have some time for questions afterwards. So joining us today are Cullen and Nick, they're going to be helping with some questions as we go through. And there's one more thing I want to hand out this form to the chat. This is the checklist that we're going to be talking about. Let's see. Did everybody get that in the chat? Okay, cool. Oh, actually, I didn't do it to everyone. Let me try my time. So this is the checklist that you can kind of follow along with as we do our presentation to kind of see what Ross is talking about as we go along. So with that, I'm going to hand it off to you, Ross, how you doing today?
Ross Tenaglia 1:08
I'm doing good. If only I could figure out how to there we go really maximize my Screen. Because I was trying to open up the the doc as well. So I know what I'm saying goes along with what they're saying. It's always helpful. But yeah, we're just all up in Vermont getting ready for some cold. So
Bret Alarcon 1:30
yeah, I don't know about the rest of the US, but man, it's gonna be cold up here.
Ross Tenaglia 1:35
Yeah, certainly. But not waste any time. Yeah, we'll get after it. I will go ahead and share my Screen. It's been a while. So I figured that one out. Everyone can see the OpenEdge explorer, right.
Bret Alarcon 1:55
Oh, yeah,
Ross Tenaglia 1:56
I had to hit Share on mine.
Bret Alarcon 1:58
There you go.
Ross Tenaglia 2:00
So is it OpenEdge? Explorer? You can see that in Chrome on the
Bret Alarcon 2:05
Yeah.
Ross Tenaglia 2:05
Okay. Excellent. And if I bring over the browser, you can see that right?
Bret Alarcon 2:10
Yes.
Ross Tenaglia 2:11
Okay, cool. So yeah, what we're gonna do is kind of talked about a few things. You know, it's going to be about, you know, checklists, performance, but it's also kind of fine tuning that ultimately does help in the grand scheme of things. So we'll be touching on, you know, both, but they're all related, and they do help. So, you know, I would assume most people here know what OpenEdge Explorer is, if you're on the RecTrac server, it's localhost not colon 9094. How you get to the back end of RecTrac, all that good stuff. Generally, Username and password or admin admin, unless you have changed it. And if you change it, make sure you know that password because it's a little bit of a pain to reset it if you forget it. So that's just something to be aware of. So in here, as you know, we have our app servers, our web speeds, our databases. And name servers are really the only other two things worry about. But generally, they're kind of you don't really think about the much. And everything else you can straight up ignore. Mainly these three collections that make or break RecTrac, literally. So you know, some of the things that we recommend, in terms of performance, for web or just in general, as well as disk space, it's all kind of lumped into the same idea is in terms of the logging, this can play a role. And that refers to the app servers, which event Demo live that your scheduled events, reports. And with the new or version of 10/5 tiene it's also going to be very important for WebTrac You know, registrations because we have made a setting change that moves receipt creation, from on demand where the web speeds for doing it. Along with doing a lot of other stuff like registrations and creating receipts, we've made a new process to offload them to the app server for events to basically share, share the load rather than having one thing do everything we split it up. Because event, scheduled events aren't nearly going to be taxed, like your web speeds are.
Ross Tenaglia 4:55
That was 1015 or newer. You And, you know, that's something to consider. And along with that, and this is what I'm going to show in event live, you can do with VIC live as well as, as the web speeds, you know, the screens are pretty identical. When you come in here, you have this configuration group. And you can do a lot of things in here. Some of it dangerous, some of it useful. One thing we always recommend, are these logging settings here, under the broker, and agent. verbose, obviously is going to make bigger logs and for event live, maybe not as important, but for client live and web live, if you're planning on being very busy, you edit this, you know, it's good. To put it on error only, that means it's not writing a lot, it's not taxing a lot. And it's creating less logging entries, so things can move a little bit quicker. And you're just removing that, generally, you put on verbose if you're like actively troubleshooting something, but if you're not, you can just throw it on error only. The so the default settings, generally is error only verbose would be something you would switch it on. So I saw you there, Keith. And then likewise, with the threshold size and max number of logs. So this right here is five megabytes, three zeros, three zeros. Five, there are some instances with there's a certain timeframe during installs where it was 500 kilobytes instead of five megabytes, those logs are pretty useless, especially if you're troubleshooting because you're literally going to get potentially minutes worth of logging throughout how many logs you have. So regardless of the circumstance, we always recommend five mega five megabytes for log, it just makes very useful log there. And the number of log files are recommend. Recommendation is 50. And so for the event live server or broker log, it will make 15 before it starts refreshing them. So 15 times five, that's how many, how much megabytes, this will take up in the logs folder. So pretty modest. If for whatever reason, you know, disk space is super premium, you can adjust how many logs you keep. But our standard is five, and 15. And that goes for the logging settings for the agents as well. Five and 15. And again, switch it to error only. You'll have less logging and you hit save. And it updates.
Cullen Barber 8:19
Question Ross Yeah, it's under load or a bigger registrations happening. Do we do what settings are recommended? They're expected error only for the for that. But any change in size or anything like that, or defaults?
Ross Tenaglia 8:34
No, the defaults are good. And I was going to pivot to the web speed I more so web live on for the next continuation of that. So yeah, error only five megabytes 15. That's norm, haven't really seen any reason for alarm with that. So with web live, you can see I jumped into it went to configuration, pretty much the same breakdown as the app servers, you can see it's very similar screens here. What's a little bit different about this, and what we're going to talk to and this can somewhat apply to RecTrac as well, but primarily more so to WebTrac, especially for big registration. Under the configuration. Here, you want to ensure your agents mine are low, they're set to 25. If you have a big registration, you have to you would want to have at least 45 here those are individual agents. That makes it so there can be up to 45 with it, and for the initial minimum you always want to set to one you don't want to say when you do a server research or anything. You don't want any app servers or web speeds to automatically kick off more than one during restart. I won't get into it. Some of you I've troubleshooted in the past before with event live that causes issues. And WebTrac likewise.
Ross Tenaglia 10:15
So we do recommend one. But then you ask, okay, well, we have a big registration coming up. And as things get busier Progress will kick off more agents, but sometimes not in a timely manner, because you're just being rushed so quickly. And the flow is faster than what the servers and the software can do. What you can do in that case is stage the amount of agents are available. So once that flood comes in, they'll be ready to pick things up real quick. And what that means is, so say, you know, you have, and this is on premise only, of course, hosted is a whole other story. But if you're on premise, what you can do, if you know that, you know, you have a registration event that begins at like 9am. And people typically start hitting the website half hour or so early, you know, to get their webpage open, what you can do is come in here and go to server pool control. And I only have one agent, because I'm my own server, so no one's hitting the website. But generally, most of you would have a handful, if people are actively using the website. But if you know, you're going to get a big rush, what you can do is submit one agent, and that lists it, we'll make another one. So you can kind of see how long that took. So you can imagine if you're being rushed with hundreds and hundreds of people. If that process is the go, I would say Naturally, there is a slight delay there. So beforehand, what you can do individually is add enough agents to start and they'll hang around for a while they won't trend themselves out, generally a half hour, you know, they'll still be kicking around. And since there's more available agents, the love would be spread amongst them, as more people are on the website. So this is one of the things we recommend, you know, in the web pre registration checklist. And this is more so the day of and this is what the process looks like you add the agents beforehand, so that they're available. And if for whatever reason how many you have aren't enough more will come but you're giving yourself enough space for the rush here.
Cullen Barber 13:07
And usually we have found that a broker like an on prem broker, single broker, the saturation point is about the 45 agents per broker. So probably don't want to go above that. I think some people have had some success at 50. But typically between 40 and 45 is the is the right mark for a single broker like that on prem. I think some people have set their minimum to 45 Ross, and restarted but it will, it will fairly quickly try to start the 45 up after you restart the broker. Right? So it's gonna,
Ross Tenaglia 13:44
yeah, you can do it that way if you restart the broker, but sometimes you don't want to restart it if you're kind of leading up to something like that. So yeah, you could do both ways, but once a little bit more preferred. So the agents don't, aren't a one to one relationship for people signing in an agent can handle multiple requests. I forgot the numbers like what one agent can handle in terms of x people. But when people are on WebTrac, depending on what they're doing, they're not making a request. They can still be on the site, but they're signing in. Yeah, that's an agent request hitting the database to log a man but once they're already logged in that agent finishes what's doing and does the next task from the next individual. And you know, a lot of it's milliseconds, and sorry, Alan made me laugh. They will die politely over time.
Ross Tenaglia 14:51
So, once things die down, they'll clear out when they're not being used. And you know, Want to make that point, you do also want to make sure that your RecTrac server has resources, you know, simple things RAM and CPU because each agent that you kick off is using CPU and using RAM. So if you're already cutting it close on day to day operations, like if you're already say, using consistently 70% CPU on average and your RAM sitting at 80%, on day to day average, you might need a bump that up on the RecTrac server for a big event, just so you have room for growth in that area. And kind of the last thing in terms of the web speed focus in here. The agent pool, so a lot of people, this won't be an issue. But you know, Nick, and I especially, you know, the case we get very often is, yeah, WebTrac is working fine. But when it gets a little bit busier, we're getting these weird web speed errors, where you hit refresh, sometimes it works. Hit refresh, again, times out with a web speed error saying agent is busy on port, XYZ, generally, it's in the 2500. range. And that's very common, because I call on this five years ago, that we increased our TCP range between the servers. So I'll just make life easy. Always just double check, you know, potentially, after this wreck chat, the bi directional TCP port traffic between the WebTrac server and your RecTrac server, you already have the main connection port, or you'd have a support case ticket with us by now. But make sure your dynamic port range of 2700 2799 is open. years ago, we expanded the range, it was in notes. But maybe that wasn't done. And if you're very busy, it may creep into a range of agent ports that are blocked. And that's a case where you'll get some people can and some people can't. So you walk away today with anything to double check. I think everyone should just double check that. Especially if you're leading up to a big registration period where you're going to be using lots of connection ports.
Bret Alarcon 17:42
Hey Ross, Joyce had a question. So for fully hosted, you mentioned this doesn't apply to us, is there anything we should be concerned about? communicated with communicating with those the team?
Ross Tenaglia 17:55
So if you have a big registration, and Cullen might be able to better answer this?
Cullen Barber 18:04
Yeah,
Ross Tenaglia 18:05
who would we who the best first touch, you could say in terms of that
Cullen Barber 18:10
it can start with support, and they usually support will contact myself or Jason Verdo? And will let ask you a few questions. And I think as Nick might have posted in one of the responses, we have a virtual waiting room, sometimes we leverage if the event is big enough, our threshold has moved up throughout this year, because we've seen some tremendous performance gains in the product since this time last year. So if there's anticipation of, you know, 900 people, you know, logging in and trying to start right at, you know, when the registration starts, we might consider putting in a virtual waiting room. So we'd like to know, you know, anticipated volume. If you've been with us for a few years, we could look at some of your historical data in your database, and some of the sessions and Google Analytics and, and receipts and things like that to help us see past history, then we lean on you a little bit to give us some thoughts on what this year could be. Are you offering more? Do you think it could be bigger because we're maybe COVID, further behind us. And that will help us decide if we want to actually have a virtual waiting room ready for you. So let's support know, and they'll we'll try to get some dates and times and stuff from you. And we'll we'll circle back with you to to find out next steps. Minimally, the hosted team will spin up some additional resources in the back end for you. Okay, that we definitely want to know. And kind of circling back a little bit. For those that are on prem, they're not hosted. We did make some pretty significant performance changes in the product. So if you have a large registration coming up, you want to minimally be on 3.1.10.15 version, minimally because that's when the big performance enhancements when into the system. So if you've got one coming up and in not on 15 or higher, definitely do that. It'll it's significant, very significant performance gains.
Ross Tenaglia 20:09
Yep. Any other questions? Brett, are we good to kind of move on?
Bret Alarcon 20:18
Additional question for on prem? Oh, they want to know, can we get the virtual waiting room?
Cullen Barber 20:25
For on prem? I guess the short answer is no. No, technically, is it possible? Yes, we use a third party product called queueit. They have a couple different interfaces. And the interface that we use with them, uses a Cloudflare software with our hosted team. So it's got some built in hooks that make it pretty simple for us to use. And so we know that and can leverage that fairly easily. I think if if you don't have a Cloud Flare option, you're going to be working with queueit kind of more directly to come up with a, you know, some Type of java script or something, you know, in that your webpage to make that work. So, potentially, there's some potential there, I just haven't heard of everybody actually doing that in the last, you know, 5,10 years. So yeah, so we don't have any, we don't have any experience currently, with anybody that's on prem using the virtual waiting room.
Ross Tenaglia 21:32
And to go with Cullen's point, you know, circling back with, you know, the performance games, the 15, you know, you know, hosted spin on those versions for a bit now. And I'm pretty sure the last few virtual waiting rooms that we've seen, didn't need them, like, we had it up and running ready to, you know, hit, hit the weight and dump everyone in the queue. But things were just going in and out quickly enough, where it didn't seem like it needed to be there. So you know, hopefully, with more and more games down the road, you know, might not even be something needed, even even for on premise, you know, especially for on premise customers. So, but with that, I'll kind of moving on, slightly, a little different here. So, you know, we talked about making changes to the, you know, you know, OpenEdge, Explorer, and so a little bit more cleanup, you can do, you know, the help things along is looking at, you know, your RecTrac, your VSI Three folder. So, on your RecTrac server, depending where you install, it might be different partition, a lot of you are probably familiar with what this you know, everything in here, because you've been here many times. So, no, in terms of cleanup, there's always a few things, you know, you have your backup folder, that, you know, you have your scheduled backup that my scheduled backup.
Ross Tenaglia 23:14
You know, I don't have one running, but you would see, you know, recent date modifieds there. And they're generally look like something like this, they have date stamps, and corresponding files. So you know, a couple of things and a couple of new things that I actually encountered here, you know, this folder, in terms of cleanup, you know, obviously, you want to have this folder as one of the more prized folders to backup the backup folder. And so if you're keeping them here, that's fine. You know, our backup cycle, generally, we have set based off of your preferences, and the backups will clear themselves out. Based off of X days, like if you have a backup Profile that says keep one backup every seven days and on the eighth day, delete the last one and overwrite it and you know, and moving forward, so you'll always have seven backups, one for each day. And then some of you kind of see this backup, you know, here it doesn't really follow the structure. This backup is typically when you do your live to Demo process when you do an on demand. Because what an on demand, all it does is at that moment, it creates a backup of the live database and then moves that backup into Demo. And that backup file doesn't get deleted. So if you have a large database, and you know disk space you want to try to clear up. When you see a backup structure that looks like this, this is always your lie. Have to Demo backup. So it's pretty safe to delete if you've already gone through that process. And if you don't do Shift Delete, if you just hit delete, it's going to you're, obviously you're recycling being. So if you truly want to get back, you got to empty that. So don't, don't forget that. And before moving to four, because I've actually, you know, with backing up the backup folder in mind, I actually don't know if they're on with me right now. But the other week, we discovered something with Carbonite, which is a pretty popular cloud based backup software. Basically, it's a glorified copy and paste system. But we did notice that they had their backup folder, backing up and due to the size of the backups, among other things, it took a very long time to Restore a Backup and to the cloud for them in their case, and that could impact performance. For one thing, if you have something like that, it could be something other than Carbonite, anything that's taxing Read Write, using your network, taking up bandwidth, if you have a large registration event or anything like that going on, and that process is running and take some time, might be good to pause it. And that will help you.
Ross Tenaglia 26:33
And another thing we discovered. If you if say the backup file, either if you running it as a backup, or sorry, restoring as a backup, or doing a on demand. If something like Carbonite is so on demand where any remote changes to files in here, it wants to start copying it. That will actually break the live to Demo process, we found the database that gets restored is corrupted, because there's two things reading it. And naturally, that's a disaster. So if you do have something like that, make sure you pause it went before you do a live to Demo. I know it's a bit of a tangent on performance, but it's kind of a new finding. And it might allude to some weird issues that we've seen over the years of outside factors and never really putting a finger on it. So that's just something to keep in mind. And along with clearing space, as well, you have your logs folder, so all the logs that you see from event live to Demo, etc, you can see all those logs are all here. And you can see mine are all roughly five megabytes. So if you have, you can sort by date modified, if you have super old logs, you know, 2022, or whatever you can, you know, come in here and you know, delete files.
Ross Tenaglia 28:17
Honestly, unless you're actively troubleshooting everything. You can pretty much delete everything. In here. There's really nothing valuable in the logs unless you're looking for something. But on day to day operations. There's really no reason to keep logs around if you don't actively need them in terms of clearing up space, or just let the amendment and you know happen as it happens. Another area that a lot of people don't know about that can take up a lot of space is under this Progress folder here is this work directory, DLC 11 Seven underscore work a lot of temp files in here, I obviously don't have a very active database. But for customers that do it will be filled with temp files, random stuff that makes no sense. And it can be very large, you can easily find out see how big it is you go to properties. You're going to have a ton of files in here. Love them small but you know the million small files is still very large. And you can have a lot of disk space to recover. And another note to depending on how your servers set up, you know the disk structure and everything. There's a lot of stuff in here. It's hard to read and open you may find yourself takes a while for this to show everything. So you got to imagine, you know might take RecTrac hard time reading and writing to it If you're experiencing a slowness, just in Windows, so what I, you know, always do in here, you can sort by date modified. And, you know, even I have a bunch of old stuff, you know, leave, you know, these folders as is but you know, go to your, you know, earliest, you know file and you can pretty much hold, Shift and delete, and start deleting the files in use, it will let you know, and you know, you have that skip, because it's already in use, and it's being written to, so I won't let you delete it. But everything else in here, for the most part is junk. And if it deletes it deletes
Cullen Barber 30:48
a couple of quick questions on that Ross came in. First one from John, the AIA VIC AIA. dot log seems to never really go down in size.
Ross Tenaglia 30:58
Yes, yeah, it will keep growing. And that's a tricky one to delete when the system is running, because while it's not being written to all the time, it's pretty darn quick. often
Cullen Barber 31:16
in use Yeah,
Ross Tenaglia 31:17
yeah. So to really delete that log, you do have to shut down Tomcat which you shut down, Tomcat will shut down VIC until you turn it back on. So that's the Apache Tomcat service.
Cullen Barber 31:34
Then another question came in. So I was asked if anybody had automated a way to kind of clean out the those work files, has anybody figured out a way to easily do without going just through manually and blasting them all?
Ross Tenaglia 31:46
I feel like I put in an enhancement. Ticket years and years ago, I don't know where it's at maybe i have to revisit it. I agree. You know, be nice. Especially if it's an update where you shut you shut down the database. So that truly means everything in here can be wiped in will be recreated as needed. I'll write that down. That's something I know I've wanted personally. And I can only imagine customers want it to, if we can consider clearing out the work directory during updates are, it would have to be during an update. It's not something that can be scheduled. Else wise. Right.
Bret Alarcon 32:33
I also want to point out, we've reached our 30 minute mark. But we'll keep going until Ross has completed everything he wants to say. If you if you need to go, I just posted a keep doing that I'm going to post a link into the chat. We'll post this RecChat By the end of the week. So if you do miss anything, you can go to that link and pick up where you left off.
Cullen Barber 33:04
And rush to get to two of the same question about Yeah, get this one a lot. Yeah. 11 and 11? Seven so can we can we wipe out all DLC? 11? No, we got to keep that because it's installed through the process installer
Ross Tenaglia 33:17
leave DLC 11. But when you see DLC 11 underscore work like Keith mentioned, you can delete that folder, the 11 underscore work, don't delete the non underscore so never touch this folder in any form. But the work one, you can delete because that just means you were on an older version of Progress years ago. And you know, we updated you guys at one point 11.7 compared to 11.3. So yes, you can delete DLC 11 underscore work. Don't touch work. Oh ye management. Yeah, don't touch this. Don't touch this, don't touch this, only this. If you're touching the other folders, generally we'd be on the phone with you on why we would ever need to go there. And just before we run on time that we already have, within this work folder, there's also this VSI folder, and RecTrac and you can delete everything in here as well. This one you could grab and delete everything. Only this stuff that are occurring today. It won't let you but this one's just temp files RecTrac You can go and clear this one, too and this one will often have a lot of disk usage as well.
Ross Tenaglia 34:59
So that oftentimes can give you a lot of disk space back. And then I have one last thing before we can open it up into, you know, more questions. For RecTrac and WebTrac server. This is a hidden disaster of disk space usage. Fortunately, nothing you can do about it because it this is Windows and ain us. But your IS logs will take up a ton of space over time, I've seen it be massive, and for certain customers, and this refers to both servers because RecTrac runs off of is WebTrac runs off of is. So your i n your eye net pub folder, this is where all your IIS logs are. And under logs. I'm locked out for some reason I noticed this a few weeks ago. But your history and logs folders, if you go in there, or you check out their properties and there. Yeah, I'm locked out for some reason, you're gonna see there's a ton of disk space being used, you can go in there and start deleting everything, because you're gonna have a log for every day. And those logs.
Ross Tenaglia 36:30
Sometimes they're 50 megabytes each, per day. So that is a secret spot where you will run out of disk space really quickly. And you're wondering, Where's it coming from the RecTrac folder hasn't changed. I've always noticed the IIS logs creep in from time to time. But yeah, and then other than that, all your other folders they are what they are not much you can do about it. They're going to be the what they take for what they take. So yeah. And, yeah, I guess you know, the last takeaway is backup your backup folders, make sure they're clean and tidy. And for the last performance thing in terms of third party software's your antivirus exceptions, you know, depending on your antivirus, it's different from different vendors can tax if it's the very dynamic that's constantly scanning, literally every interaction? That could certainly slow down. Our RecTrac operates? Or, you know, if you put on the new antivirus, and it completely nukes RecTrac That could be why to in terms of making it not run because it's running against it all the time, then you really know it's a problem. So yeah, I guess at this, I will kind of leave it as that at that I know there's you know, a million things in the checklist below. Those are within RecTrac. So we don't often have the opportunity to cover the ones that are outside of RecTrac server side related. So you know, that was one of our points to cover those aspects of it that are often foreign and are really often times the phone calls we get, hey, I did all this and RecTrac I'm unfamiliar with the server side aspect of it. So this is nice to show. So you can see that end of the spectrum.
Bret Alarcon 38:48
Hey, oh, thank you, Ross. We've been doing a pretty good job of picking up questions as they come in. But we'll leave a couple more minutes open. In case anybody has any other questions. Frank says good info. Thank you, Ross. So yeah, as I said before, I'm going to be posting this RecChat The video to our support site. I'm also going to be including the two documents as well, I'll repost those just in case people got in a little bit late. And they weren't able to see that in the chat. So if you look in your chat, you should see our two checklists, documents there as well. As he has a general question, are there any other solutions for event timer stopping?
Ross Tenaglia 39:38
That, that yeah, that certainly is a loaded one. There are many reasons why we see it spontaneously stopping and every time it can be different. I dealt with a case it was because they're like Hey, it's stopping every time at 3am Every day, it's crashing at 3am. And we had to restart it works all day. Next morning, 3am. Lo and behold, someone scheduled, literally 180 reports to run exactly at 3am. I don't even think our hosted servers can handle that, let alone any sort of on premise or even the software in general. And we found out that was crashing it. So you know, they made sure to stagger their scheduled reports. So maybe 1015 reports per five minute block, you know, eventually went all its way until like, probably 5am to cover all those reports. But that was a reason why he's doing that. Another thing could be disk space, this happens a lot, you know, crashes at a particular time, always well, that's when their backups run. And if, if it's trying to run the backup, and you run out of your disk space, the event timer will crash. Because it basically gets jammed up, think of like a pile up. That's, you know, they're really the easiest way to give you an analogy. RecTrac may pseudo work, but eventually it'll just outright crash. Because of disk space. You know, that's a common one. Beyond that, you know, sometimes we had to get in the weeds, there could be you know, someone ran a bad report with an outrageous query, and or had bad data. You know, the list goes on. But there is always some and sometimes there's a third party software is messing up the event timer running or the or just straight up, the service doesn't start. It's a tricky one. I know a lot of people have questions on it. Sometimes there's not a catch all answer and requires us to suss out with you. But we try to make sure we figure it out.
Bret Alarcon 42:00
All right. Well, I don't see anything else. Oh, questions? How often? Do we need to restart the AppServer? When we make changes to the configuration?
Ross Tenaglia 42:10
If you change the amount of maximum agents? Pretty sure, yes, yeah. You can't just trim it. If you change the max agents, you have to do a quick restart. So, you know, you might want to do that beforehand. You know, if you do a quick restart, you know, it depends on if it's client live or web live, if you know, there's not a lot of people on there, shutting it off, turn it back on, you can literally do it all within 30 seconds. You just have to gauge you know, not a lot of people are using it or on it. And if they are, as long as they're not moving around, they won't notice anything. It'll catch itself, you know, right back up. Yeah, that's app servers, web speeds. When you make those changes, that's what happens.
Bret Alarcon 43:00
All right. Well, that looks like it's it for questions. Thank you, Ross, for joining us today and giving us all the information. And thank you everybody out there for attending. Good session. Yes, Keith says, as always, thank you very much. And with that, we'll leave you guys to the rest of your day. I hope you enjoy your day and have a good one to
Cullen Barber 43:22
Bye everybody.
Ross Tenaglia 43:23
Bye, everyone.