Thanks for the EtreCheck report.
Short answer
The report's drive results tells a very clear story. Your iMac is equipped with a Fusion Drive, which consists of a smaller capacity SSD and a much larger capacity HDD. The report is telling us that the SSD tier looks perfectly healthy, while the HDD tier is the more likely source of slow responsiveness, even though it isn’t yet throwing explicit SMART errors.
A bit longer answer
- The SSD (disk0) is behaving normally. It has 1% wear, a typical write-total for its age, and “100% health.” The unsafe shutdown count is high, but that’s common on Fusion setups and doesn’t indicate failure. The read/write speeds EtreCheck measured align perfectly with the SSD tier’s NVMe performance. Nothing about disk0 suggests trouble.
- The HDD (disk1 — ST1000DM003) is the weak point in most late-model iMac Fusion Drives. Seagate’s ST1000DM003 is known for:
- High latency when waking from idle
- Slower-than-average seek times as they age
- Occasional long pauses under APFS workloads, especially once the drive gets past ~50% full. Your drive is sitting around 645 GB used, meaning the HDD tier sees a lot of action, and APFS is not particularly kind to spinning disks.
- The “File system: 19.72 seconds” delay is the only true red flag. This number tells you that at the moment EtreCheck probed the disk, the filesystem took nearly 20 seconds to respond. When this happens on a Fusion Drive, it’s almost always the mechanical disk stalling — not outright failure, but a sign of:
- Slow waking or seeking
- Fragmented/densely-packed APFS container space
- Early mechanical sluggishness
If the HDD were *actually failing*, you’d expect slow HDD-tier benchmarks (<100 MB/s), kernel I/O errors, or SMART warnings — none of which are showing up in the report.
So, based on what you’ve provided, this looks more like a typical aging Fusion Drive scenario rather than an imminent hardware failure. The SSD tier is strong, and the HDD tier is simply behaving like a 5–6 year old Seagate spinner under APFS.
A common solution in these cases, where you still want to keep your aging iMac, but get better performance, has two basic options:
- Replace the Fusion drive with a single large capacity SSD, or
- Boot up your Mac from an external USB-C/SATA SSD.
Moving on the the reports' Diagnostics section:
Those diagnostics fit the broader pattern we were already circling: nothing screams “hardware failure,” but the system is clearly under strain, and most of that pressure points to low RAM (8 GB) and the Fusion Drive’s mechanical tier rather than catastrophic disk issues.
Let's look at each of the items in this section and what they could indicate.
- `knowledgeconstructiond` and `spotlightknowledged` spinning up - These two processes belong to Apple’s machine-learning and indexing subsystems. On systems with limited RAM and slower storage, they tend to flare up:
- When Spotlight reindexes
- When Photos, Mail, or Finder metadata is being processed
- When the system is low on free memory and starts thrashing the swapfile
On an 8 GB Fusion Drive iMac, this is extremely common because macOS leans on virtual memory heavily. The HDD tier is terrible for swap, so any indexing or ML background tasks get bottlenecked and spike CPU.
- `MTLCompilerService` crashing - This one usually points to:
- GPU driver quirks
- An app invoking Metal shader compilation (games, browsers, Zoom/Teams, certain creative apps)
- Low memory or memory pressure during those tasks
A few isolated crashes here aren’t unusual on Catalina/Big Sur/Monterey-era systems, particularly with just 8 GB of RAM. It’s rarely harmful beyond causing a brief app hiccup.
When you combine the diagnostics with the Fusion Drive's slow HDD responsiveness and the 8 GB RAM limit, the user’s symptoms likely show up as:
- Beachballing
- Slow app launches
- Delayed wake-ups
- Finder hangs
- Safari or Chrome tabs refreshing unexpectedly
Nothing in the diagnostic section suggests imminent hardware failure. What it does suggest is a system that’s being pushed beyond what a Fusion Drive + 8GB can comfortably handle in 2025 workloads.
