Granting Administrator Privileges to users

Granting Administrator Privileges

Granting administrator privileges to users is one thing that I often see, that is done much more than it should be. The reason given is this simplifies those occasions when software needs to be installed or updated. However, in general, this is a bad practice.

The simplification of installing legitimate software will be exploited if you are ever the victim of malware or viruses. Since the user that the infection uses already has administrator privileges, the malware does not need to find anything to exploit for it to have access to your entire system. If the malware already has inherited the privilege from the user login and can wreak whatever havoc is in its payload.

On the other hand, if the user is given basic user privileges, and the malware is unable to find any way to exploit the system to increase its privileges, then the damage is contained to what the captured user can modify. While that is still upsetting, having to restore some documents, and change user settings is more desirable than having the entire computer system corrupted and compromised. Not to mention, the potential effect on other computers on your network, if administrator-level privileges can be leveraged against them.

With properly set permissions, users can do everything that they need, without elevated privileges. It can take some extra time to setup the appropriate permissions, but not nearly the time that it takes to recover from a malware attack that could have been limited in scope by a simple restriction of administrator level privileges.

Dave

Signs Your System Has Been Compromised

Signs Your System Has Been Compromised

Users need to be on the alert for signs that your system has been compromised. Let us look at some of the signs. One that often occurs is that icons on the desktop have been moved, or new icons have appeared. If new icons have appeared, and no software has been installed, as far as is known, it is a red flag that the system has been compromised.

Hackers will often install new software when they gain access to a system, to give them further control, or the ability to gather additional information from the system. Most times, there is no obvious trace of this additional software, but many times they are sloppy, and those surprising new icons are an indication.

Icons being moved is often also an indication. This is often due to the hacker going through and clearing those new icons. However, those icons moved some of your old icons when they were installed, and the hacker does not know exactly where the old icons were on the desktop. Or, the icons were subsequently auto-arranged, and are thus not in the location on the desktop that users had moved them to.

Another big indicator is that your system suddenly slows down. The screen may lag when you are typing, or processes that used to take a second or two, now take many seconds, to even minutes. This is due to the system having an increased load, due to the additional software that has been installed, and what it is doing to steal information or use the computing power for other nefarious activity, such as cryptocurrency mining. Cryptocurrency mining is using computing power to unlock cyber currency, such as Bitcoin. This takes a great deal of computing power, cybercriminals are farming this activity out to compromised systems, rather than using their own.

Basically, be alert to any change in the way that your system looks, or responds. While it may be due to legitimate changes, often it is a sign that unwanted people have gained access to your system.

CCS provides IDS products.  Contact our Sales Department to discuss the right solution for your operations.

Dave.

The threat of Attack on Point-Of-Sale (POS) systems

The threat of Attack on Point-Of-Sale (POS) systems

In the retail world, the threat of attack on Point-Of-Sale (POS) systems is always high, as they are a prime target for the bad actors trying to get credit card information. There is always new malware being created to try to get that information. There are, however, steps that can be taken to reduce the risk, even with regards to the latest attack software.

The best thing that you can do is to isolate your servers and workstations as much as possible. In an ideal world, these would be completely isolated, and not interact with any other systems. The reality is that this is very rarely the case. With that in mind, let’s look at some things that can be done.

The first step is to make sure that your firewall is as restrictive as possible. Your POS workstations and server if needed should only be restricted to accessing what is absolutely needed. In most cases, this is only your credit card processor. In no case, should your firewall allow either the workstations or the server, unrestricted access to the internet? The firewall will eliminate the chance for drive-by downloads, or users browsing to malicious sites.

Along those lines, any computers that are used for general internet browsing, and also email, should be on a separate network. If it is not possible to implement workstations on a separate physical network, at least use a different logical network as in a different network address range. Even just the different logical network, will stop the majority of malware infections.

If remote access is needed, and it should be restricted to those cases where it truly is needed such as your support company, then it should be restricted to only those addresses that have a legitimate reason to connect. Also, any such connections should be closely monitored. One such method is to disable the remote access software, and only enable it when your support personnel is actively connecting. Of course, it is again disabled as soon as they are finished.

Those cover the basics of securing your POS system. It is a good place to start, but it is only a start. Keeping software updated, training personnel, and keeping vigilant are always key components, also.

Dave

Keeping Systems up-to-date

Keeping Systems up-to-date.

Everyone should know the importance of keeping your operating system, and software, up-to-date. Even with the amount of information on the subject, there are still many that allow this task to slide for excessively long periods of time. That is a very, very, bad thing to do, given the volume of malicious software that we are constantly bombarded with.

While I am not personally a fan of automatic updates, that is a far better choice than not installing updates at all. If you find that you are not taking the time to apply the patches and updates, then, by all means, configure automatic updates, and let the system do it for you.

While automatic updates involve an amount of risk, it is much less than the risk run by not having current fixes and patches applied. My preference is to periodically (and frequently) manually install the updates. Manually installing allows me to check for known issues regarding those updates, and make the decision to install a particular patch, or not. This approach does take some dedication on your part, though, and as such is not for everyone.

Whichever method you choose, be sure that your systems are updated with the current patches and fixes. Otherwise, you are running the very big risk of waking up to a nasty situation, in which your computers have been compromised, or are being held for ransom.

Dave.

Testing Updates and Changes

Testing updates and Changes.

With the recent changes in requirements for credit card processors, NCR Retail Online going away, and others, there are a large number of people going through updates and changes. Many of these changes are quite significant. With that in mind, a word about testing those changes is in order.

The most common thing that I see is a lack of sufficient testing. For example, if you are changing your web store provider, there will be changes in how orders flow in and out of your ticketing system. It is not enough to test that orders come into your system. You also need to check that you can process those orders. It does not do any good to receive orders, and then not be able to process and complete them. The testing process needs to incorporate all aspects of the workflow.

  • Does the order come in with the correct information?
  • How about taxes?
  • Are the tax amounts correct when appropriate?
  • How about when taxes do not apply so that there is a zero
  • tax amount?

Those are all important, but that is just the beginning.

If that information is importing correctly, what about going forward.

You should test printing picking tickets, generating invoices, and everything all the way through posting those invoices. When that all goes through without a hitch, there is still more to do. Test canceling orders, and partial shipments. If you ever edit orders to add or remove lines, change a line quantity, or anything else, then those functions need to be tested also.

It takes some time and needs to be included in your schedule for everything else, but the more time spent testing changes and updates ahead of time, the smoother it will be when it comes time to implement those changes in your live system.

Dave

Recalculating Inventory in a Counterpoint Multi-site Environment

Recalculating Inventory in a Counterpoint Multi-site Environment

If you are using Counterpoint in a multi-site environment, care must be taken when it comes to running the Inventory Recalculate Quantities procedure. This procedure checks several of the quantity fields in the inventory tables, against the data in the rest of the tables. Fields such as quantity-committed, quantity-on-po, and quantity-on-transfers, among others, are checked and corrected if necessary. Note, that it will not adjust quantity-on-hand. That must be done via inventory adjustments.

The issue in a multi-site environment relates to how replication works. When a replication session starts with a remote site, it does not take a snapshot of the database. Rather, it goes through the list of tables and makes the appropriate updates as it processes the various tables. What this means as to the inventory quantity recalculations, is that after finishing replication, all of the data is not consistent as to the point in time that it represents, due to ongoing user activity.

To illustrate, let us look at the purchasing process and inventory quantities. The purchasing tables are processed by replication in the first half of the list of tables, while the inventory tables themselves are processed nearly at the end. This process leads to the possibility that a replication session will start, and get through the purchasing tables. Then while replication processes other tables, a user at the remote site posts a new purchase order. A bit later, the replication processes the inventory tables. In this case, after the replication is finished, the hub site has no records of the purchase order that was posted at the remote site. However, it does have updated inventory records showing a quantity-on-po. In the Inventory Recalculate Quantities is run at the hub at that point, the result would be to set the quantity-on-po to zero for all items on the purchase order that was posted (assuming that they are not on any other purchase orders). When replication runs again to the remote site, the purchase order will then be transferred to the hub. Also, the quantity-on-po that was set to zero at the hub will be sent to the remote. Now both the remote and the hub have an open purchase order, while the quantity-on-po for all of the items on that purchase order will be zero.

Inventory quantities should only be recalculated after replicating with the remote sites when no activity is occurring at those sites for the entire duration of the replication, avoiding this timing problem. Typically, this means that the recalculation must be run well after the remote site closes, or before they begin processing in the morning. Remember, it is not enough that the remote site has closed, and everyone has gone home. A replication must occur after they have closed and gone home. Only then will the recalculation routine have data that truly reflects the state of the data at the remote site. Most of the time, this means that the best time to run the recalculation is in the morning before users at the remote site start doing anything with Counterpoint.

Dave.

Recovering From a Ransomware Attack

Recovering From a Ransomware Attack

Ransomware attacks are on the rise. It is getting more common to get random emails with subjects indicating they are package tracking, voice mails, photo edits, and so on. Many of these are attempts to get you to take the bait, click the link, and ultimately install ransomware. Much has been written about recognizing malicious emails, not opening mail from unknown users, and other good advice. What happens, however, if you are unfortunate and do get hit with ransomware?

We are assuming this is a true Ransomware infection, where an active payload of malware has been added to your system. Another type of Ransomware is Scareware masquerading as Ransomware. This latter Ransomware has no payload but threatens you with data encryption as well. It is best to assume any Ransomware threat includes a payload, at first. A safe mode reboot investigation can help you check if a payload is active. If the infection is just Scareware, you may be safe with a reboot and comprehensive malware scan to confirm there is no infection active.

First, be very suspicious of any unusual activity. One of the first signs, even before the ransom notice pops up, is that programs will stop working, or documents will disappear. This activity is due to the malicious software starting to encrypt your files. If anything like that happens, take immediate action. First, disconnect your computer from your network. That is, physically disconnect the network cable, or if you use a wireless connection, turn it off. Also, immediately shut down your computer. I do not usually advocate just turning off the power, but this is one time that it is not a bad idea. The idea is that if ransomware has started on your system, to limit the damage occurring.

Try to start your computer in safe mode, and begin investigating. Make sure you lookup entering safe mode in Windows on your version to MAKE SURE you do NOT get a normal boot or the Ransomware will be active again. Check for those programs or documents that suddenly disappeared. If there is a file with the same name, but the extension has changed, most likely ransomware is the culprit. In that case, be prepared to do some research, and possibly still lose some work. It depends on the active Ransomware variant since some have been Ransomware payloads have been cracked and there are recover utilities available.

Other Ransomware payloads do not have removal utilities, and you will have to go to your back copies. Before that, however, you need to make sure that the machine is cleaned of the ransomware programs or your system will be reinfected and you will need to start over again. If utilities exist to clean the Ransomeware for your system, they should be used immediately. If not, a lot of digging and experimenting will be required. If there is no cleaning utility you may need to reformat the infected drives, reinstall the operating system, and then restore from a full image backup, NOT just a file backup. In either case, spend a lot of time checking your system, before putting it back on your network and getting on with your work. You want to be very, very, sure that the Ransomware is gone, or you will be exposing the rest of the computers on your network to Ransomware infections.

Recovering from Ransomware is a critical task that can be very complex.  This blog is just a simple overview.  We recommend you contact CCS Retail Systems Support for further guidance and services to ensure the Ransomware is properly eliminated from your systems. Remember that if you comply with the Ransomware demands there is NO guarantee that your payment will result in any recovery of your system.  The best course of action is to defeat the Ransomware request NOT honor it.

Dave.

Automatically Applying Windows Updates on Servers.

Automatically Applying Windows Updates on Servers.

It is important to apply updates to your Windows server. These updates help to keep your system secure and running at peak performance.

I do not recommend automatically applying these updates on your server, though. It is much better to set Windows updates to either “notify”, or “download updates and notify”. The biggest problem with automatically installing updates is having the server reboot. This may occur automatically. Even if the time for installing updates is set to sometime in the middle of the night, when you are off the system, such a reboot may occur later, when you are on. I have seen this in cases where the updates took a long time to install. Also, however, the reboot may be delayed. If for example, a user is left logged in on the console, the reboot may be delayed until after the user is logged off. Of course, that would typically occur when you are open and using your server. Also, some updates are installed when shutting down or starting up. I have seen such updates take an hour, or more, to install. Your server is unavailable during that time.

In addition to inconvenient restarting of your server, occasionally an update may require additional attention. Although it is rare, there may be recommended setting changes, or other manual steps. It is best to know about these and be prepared to address them, instead of suddenly being confronted with a server that is not working as desired.

My recommendation, therefore, is to not automatically apply updates on your servers. Instead, make it a frequent task to check the updates that are released, and determine which should be applied. Then, after checking for any possible issues with your setup, or software, to install those updates at a time when the server may be rebooted if needed. Then check your system after they are installed, and rebooted if needed, to make sure your system and software are operating properly.

While that is my recommendation for servers, it applies to a lesser degree to your workstations, also. For workstations, you should determine if you can tolerate an unanticipated reboot, with some possible downtime. If resources permit, I recommend handling your workstations the same as your servers, and reserve updates for manual processing.

Dave.

Purging Distributions

Purging Distributions

As you use your Counterpoint SQL system, you will accumulate data. Sales history, purchasing, and receiving, all of the history is accumulated, and these tables can grow to be quite large.

One table that can grow very fast, is the distributions table. Everything that is posted will create distributions. Depending on the settings used, it can be many records for a single posting. In a short time, this can build into thousands, or millions, of records. Moreover, most people do not need most of the data in this table. If your interface to another accounting system, once the distributions are sent to that other system, one really has no need for them in Counterpoint. Also, many do not use the accounting function, either within Counterpoint, or interfaced to another accounting system. In that case, those records are never used and are just taking up space.

Fortunately, distributions are one of the areas that Counterpoint has a purge function. From the main menu, it is under System -> Accounting -> Utilities -> Purge Distributions. Under the event tab, select the Post Date range that you want to purge. You may have to use the customize function to find the post date option. Make sure that you select a range that will leave any records that you may want. In a lot of cases, leaving 30 days of distributions is appropriate. However, that will vary, depending on your use of these records, and your needs. You may be able to purge all records (for example, if you do not use any of the distribution information within Counterpoint).

Once you have selected the purge range and selected the purge option, you will get a report of the records that will be purged. Also, you will be asked for confirmation that you actually wish to purge the records. Once you confirm that you want to purge, the records will be removed. This can take several minutes, depending on how many records you are purging at a time. That is all there is to it.

Dave.

Getting ready to upgrade Counterpoint SQL

Getting ready to upgrade Counterpoint SQL

With many people getting ready to upgrade Counterpoint SQL to the latest version, I thought I would talk about what can be done to prepare. One needs to confirm that the systems have the required prerequisites, as well as clearing any data that can be.

On the systems side, the biggest requirement is that SQL Server be version 2008 R2, at the minimum. Version 2012 is also supported, but later versions are not recommended. They may work without issues, but we are not recommending using versions above 2012 at this point. If you need to upgrade your SQL Server software, you can purchase a 2016 license, and install 2012 as a downgraded install, using that license. That way, you can upgrade to 2016 under the same license should it be necessary, or desirable, at some point in the future. That is the only issue that we have been seeing as far as system requirements for upgrading Counterpoint SQL.

As to the data, purging old data will speed up the upgrade process. In particular, purging the ticket history can make a big difference. Many times purging old ticket history is something that has not been done in a long time, if ever. With many interconnected tables making up the ticket history, the number of records that even a small store can generate overs years can be amazing. During the upgrade of Counterpoint SQL, the ticket history tables will updated with new fields. On very large history tables, this can be very time consuming.

In order to lessen the impact, the upgrade process does not perform the actual ticket history table upgrades. Rather, the tables are move, and empty tables with the changes to the schema are created. These tables are then populated from the data in the old tables, by means of a scheduled job. This job will migrate a number of records each time that it runs. In most cases, it will be scheduled to run frequently, and process a fairly small number of history records, starting with the latest tickets. Thus, over time the entire data set is migrated, and the records most needed for functions such as purchasing based on sales, or replenishment, are migrated first. Then, older tickets, which are usually not needed as much, are migrated later.

By purging the ticket history prior to upgrading, the migration process will obviously complete sooner. Every company is different in what they need for history, so some time should be spent in determining how much history to preserve. Two to three years seems to be the most common, but some wish to retain more, and some less. Once the cutoff date is determined, then the history should be purged up to that date. However, if you are using multi-site, I strongly recommend purging only one or two months of data at a time, and letting those changes replicate to all sites. If several months, or years of data are purged at once, it can severely impact the replication and even bring it down.

Those are the main points to address in preparing to upgrade Counterpoint SQL. Once they are done, it is simply a matter of scheduling a time and doing the upgrade, which usually takes a few hours at each location.

Dave.