When "idle" isn't idle: how a Linux kernel optimization became a QUIC bug

The article describes a bug in Cloudflare's QUIC implementation (quiche) where the CUBIC congestion control algorithm's congestion window (cwnd) becomes permanently stuck at its minimum value after a period of heavy packet loss, preventing recovery. The bug was triggered by porting a Linux kernel optimization from RFC 9438 to QUIC, causing approximately 60% of integration tests to fail when 30% random packet loss was applied during the first two seconds of a connection. The issue was resolved with a simple, near-one-line code fix that broke the cycle of the cwnd remaining pinned at its minimum.

CUBIC, standardized in RFC 9438 , is the default congestion controller in Linux, and as a result governs how most TCP and QUIC connections on the public Internet probe for available bandwidth, back off when they detect loss, and recover afterward. At Cloudflare, our open-source implementation of QUIC, https://github.com/cloudflare/quiche quiche , uses CUBIC as its default congestion controller, meaning this code is in the critical path for a significant share of the traffic we serve. In this post, we’ll tell the story of a bug in which CUBIC's congestion window cwnd gets permanently pinned at its minimum and never recovers from a congestion collapse event. The story starts with a Linux kernel change aimed at bringing CUBIC into line with the app-limited exclusion described in RFC 9438 §4.2-12 — a fix to a real problem in TCP that, when ported to our QUIC implementation, surfaced unexpected behaviors in quiche. It has a happy ending: an elegant near- one-line fix that broke the cycle. CUBIC's logic in a nutshell Before we dive into the core problem, a quick refresher on Congestion Control Algorithms CCAs may help to set the stage. The central knob a CCA turns is the congestion window cwnd : the sender-side cap on how many bytes can be in flight sent but not yet acknowledged at any moment. A larger cwnd lets the sender push more data per round trip; a smaller cwnd throttles it. Every loss-based CCA, CUBIC included, is ultimately a policy for how to grow cwnd when the network looks healthy and how to shrink it when it doesn't. In essence, CCAs aim to maximize data transfer by inferring the "available bandwidth" of the network; because no one wants to pay for a 1 Gbps subscription and only use a fraction of it. The family of loss-based algorithms, to which CUBIC belongs, operate on a fundamental premise: 1 if there is no packet loss, increase the sending rate i.e. increase the bandwidth utilization ; 2 if there is loss, loss-based algorithms assume that the network's capacity has been exceeded, and the sender must back off i.e. decrease the bandwidth utilization . This logic is built on several assumptions that have been revisited over the years. However, we'll save that discussion for another time. The symptom: a test that fails 61% of the time Our investigation started with the report of unexpected failures in our ingress proxy integration test pipeline. This erratic behavior appeared in tests where CUBIC was evaluated in a scenario of heavy loss in the early part of the connection. Recovery after congestion collapse is an uncommon regime, but it is exactly the regime a congestion controller exists to handle. Most congestion control tests exercise the steady-state and growth phases of an algorithm; far fewer probe what happens at minimum cwnd, after the connection has been beaten down. Bugs in this corner of the state space are invisible in throughput dashboards, undetectable by static review, and only surface when you deliberately drive a CCA into it and watch whether it can climb back out — which is exactly what this test did. The simulated test setup includes the following details: Quiche HTTP/3 client and server running at locally localhost RTT = 10ms set up in the configuration A 10 MB file download over HTTP/3 Using CUBIC congestion control With 30% random packet loss injected during the first two seconds After two seconds, loss stops entirely The test has a generous 10-second timeout to complete the download, which is expected to be completed in four or five seconds The expected behavior is straightforward: CUBIC should take some hits during the loss phase, reduce its congestion window, and once loss stops, steadily ramp up and finish the download well within the timeout. Instead, we observed in multiple 100-time runs that around 60% of our tests were not able to complete the download within the generous 10-second timeout. The anomaly: 999 state transitions with zero loss We instrumented https://github.com/cloudflare/quiche quiche's qlog output with packet loss events and built visualizations to understand what was happening inside the congestion controller: Connection overview of a failing test. After T=2s, packet loss stops entirely — yet cwnd remains pinned at the minimum floor and the congestion state oscillates between recovery and congestion avoidance every ~14ms. After the two-second 2000 ms mark, packet loss stops entirely. However, the number of bytes in flight remains flat, which contradicts the core logic of the CUBIC algorithm: in the absence of loss, apply more gas to increase throttle more bytes in our world . This raises the question: if the network is no longer dropping packets, why is the congestion window failing to grow? When we zoom into that region, our analysis shows that CUBIC enters a rapid oscillation, shown in our plot as an extended recovery phase, between congestion avoidance state the operational regime phase and recovery state the packet loss recovery state — 999 transitions in approximately 6.7 seconds. That’s one transition every ~14ms — suspiciously close to the connection's RTT 10ms . Throughout this entire period, cwnd is locked at the minimum floor: 2700 bytes, or two full-size packets. Clearly something in CUBIC's logic is misinterpreting the state of the connection. The key clue is the oscillation period: ~14ms matches the RTT. Whatever is triggering the recovery/avoidance flip is happening once per round trip, in lockstep with connection's ACK clock; the self-clocking rhythm in which each round-trip's ACKs from the client trigger the server's next send. Because this is a download server to client , the ACKs in question travel client to server, and CUBIC's state machine runs on the server side: every time those ACKs land, bytes in flight drops to zero and the server sends the next two-packet burst, which is what triggers the bug. To confirm this behavior was CUBIC-specific, we ran the same test with Reno , another member of the loss-based family but with a different growth rate. The results were conclusive: 100% pass rate, showing Reno recovered cleanly after the loss phase, and revealing that this is a CUBIC-related bug. Reno recovers cleanly after the loss phase ends at T=2s and completes the download by ~5s Loss-based algorithms have two pedals, gas and brake, with a difference in how they accelerate. Well, CUBIC comes with some extra features. Here we are going to focus on bytes in flight == 0. TCP CUBIC after idle Linux, 2017 To understand the bug, we first need to understand the optimization it came from. In 2017,an issue was found with Linux kernel's CUBIC implementation. The https://github.com/torvalds/linux/commit/30927520dbae297182990bb21d08762bcc35ce1d commit message explains: The epoch is only updated/reset initially and when experiencing losses. The delta "t" of now - epoch start can be arbitrary large after app idle as well as the bic target . Consequentially the slope inverse of ca- cnt would be really large, and eventually ca- cnt would be lower-bounded in the end to 2 to have delayed-ACK slow-start behavior. This particularly shows up when slow start after idle is disabled as a dangerous cwnd inflation 1.5 x RTT after few seconds of idle time. The epoch is the reference timestamp CUBIC uses to anchor its growth curve: W cubic delta t is parameterized by delta t = now - epoch start , and the epoch is reset whenever CUBIC restarts its growth function — most notably after a loss event reduces cwnd . Between resets, delta t grows monotonically with wall-clock time. When an application goes idle stops sending for a while and then resumes, the CUBIC growth function W cubic delta t computes delta t as now - epoch start , as illustrated in the figure below. Since the epoch wasn't updated during idle, delta t is huge, producing an enormous target window — and CUBIC would immediately try to inflate cwnd to an unreasonable value. Jana Iyengar's initial fix was to reset epoch start when the application resumes sending. But Neal Cardwell pointed out the flaw in that approach: …it would ask the CUBIC algorithm to recalculate the curve so that we again start growing steeply upward from where cwnd is now as CUBIC does just after a loss . Ideally we'd want the cwnd growth curve to be the same shape, just shifted later in time by the amount of the idle period. The elegant solution, authored by Eric Dumazet, Yuchung Cheng, and Neal Cardwell, was to shift the epoch forward by the idle duration rather than resetting it. This preserves the shape of the CUBIC growth curve — just sliding it in time so that the algorithm picks up where it left off. The port to quiche 2020 When CUBIC was first implemented in quiche, this idle-period adjustment was ported. However, QUIC, which runs in the user space, doesn't have TCP's kernel-level CA EVENT TX START callback. Instead, the quiche implementation checks for the idle condition inside on packet sent : // cubic.rs — on packet sent simplified /// Updates the state when a packet is sent. fn on packet sent &mut self, bytes in flight: usize, now: Instant, ... { // If the sending burst is restarting i.e., bytes in flight was zero before this send , // adjust the congestion recovery start time to account for the gap in sending. if bytes in flight == 0 { let delta = now - self.last sent time; self.congestion recovery start time += delta; } // Record the time of this send event. self.last sent time = now; } Where it breaks: the QUIC difference The fix ported to quiche included a bug in the original kernel change which was fixed by a followup change to the kernel cubic module about a week later. The commit message for the second fix explains: tcp cubic : do not set epoch start in the future Tracking idle time in bictcp cwnd event is imprecise, as epoch start is normally set at ACK processing time, not at send time. Doing a proper fix would need to add an additional state variable, and does not seem worth the trouble, given CUBIC bug has been there forever before Jana noticed it. Let's simply not set epoch start in the future, otherwise bictcp update could overflow and CUBIC would again grow cwnd too fast. As mentioned in the commit message, recovery start time is set during ACK processing, and the computation of the adjustment based on sent times can push the recovery start time into the future. This explains the oscillation between recovery and congestion avoidance seen on our test. The trap only consistently triggers when every incoming ACK drives bytes in flight all the way to zero — which in practice means cwnd has collapsed to its minimum two packets and the application has data ready to send another full window the moment an ACK arrives. Outside this regime, bytes in flight == 0 is less likely to hold on every send, so it is less likely to trigger the bug. Why doesn't this also happen at connection start? The bug only triggers when the connection exits slow-start and switches over to congestion avoidance. Before exiting slow-start, congestion recovery start time is not set, so the buggy branch in on packet sent has no recovery boundary to advance. During slow start CUBIC's cwnd grows by the same Reno-style ack-based rule shared by all loss-based CCAs — the cubic curve and its sensitivity to congestion recovery start time only enter the picture once the connection is in congestion avoidance, meaning the trap needs three things at once: a real loss event to set the recovery boundary, congestion avoidance to be running, and cwnd collapsed to the two-packet floor. The self-perpetuating recovery trap. At minimum cwnd, every ACK cycle triggers the idle period adjustment with an inflated delta. At a minimum cwnd two packets , the dynamics of the connection shift into a "death spiral" where the idle period optimization becomes a self-fulfilling prophecy. This trap operates in a continuous loop: Send and ACK packets: The sender transmits the entire two-packet window. After one RTT ~14ms , both packets are ACKed, causing bytes in flight to drop to zero. False idle detection: When the next burst is sent, on packet sent sees bytes in flight == 0 and assumes the connection was idle, but it was congestion limited. Inflated delta: The calculation uses now - last sent time to determine the idle duration. When the congestion window cwnd is at its minimum, last sent time is the timestamp of the start of the previous RTT cycle. Therefore, the resulting delta is approximately 14ms the connection's RTT + additional rounding errors . This RTT-sized delta is incorrectly applied as the "idle" time. The actual time the connection was idle the processing gap between the last ACK arriving and the next packet being sent is effectively 0. By measuring the full RTT instead of the true gap, the delta is inflated significantly, aggressively shifting the recovery start time forward, possibly into the future. Perceived recovery: Because the recovery start time is now in the future, the in congestion recovery check returns true for every incoming ACK. Processing of the next ACK exits recovery and sets the recovery start to the ACK time which is larger than last sent time, making it likely for the congestion controller to push the recovery time into the future when doing the next send. Stagnation: Since CUBIC skips cwnd growth for any packet perceived to be in a recovery period, the window remains pinned at two packets — ensuring the pipe drains completely on the next ACK and restarting the cycle. And this loop repeats for thousands of cycles until the accumulation of small deviations — from scheduler jitter and ACK processing variance — lets the <= boundary in in congestion recovery slip behind the next packet's send time, breaking the cycle. The fix: measuring idle from the right moment Fixing the death spiral involves measuring the idle duration from when bytes in flight actually transitioned to zero the last ACK processed rather than the last packet sent. Add last ack time timestamp to the CUBIC state. Update that timestamp when ACKs arrive. Use it for the idle delta computation: // cubic.rs — on packet sent fn on packet sent &mut self, bytes in flight: usize, now: Instant, ... { // Check if the connection was idle before this packet was sent. if bytes in flight == 0 { if let Some recovery start time = r.congestion recovery start time { // Measure idle from the most recent activity: either the // last ACK approximating when bif hit 0 or the last data // send, whichever is later. Using last sent time alone // would inflate the delta by a full RTT when cwnd is small // and bif transiently hits 0 between ACK and send. let idle start = cmp::max cubic.last ack time, cubic.last sent time ; if let Some idle start = idle start { if idle start < now { let delta = now - idle start; r.congestion recovery start time = Some recovery start time + delta ; } } } } With the delta now reflecting the actual gap since the last ACK, the recovery boundary stops chasing the send time: Old code: boundary advances one RTT per cycle, always landing on or ahead of the next send. Fix: boundary barely moves; the next send lands ahead of it and cwnd grows. For genuinely idle connections, last ack time is far in the past and the same expression captures the full idle duration, the original epoch-shift behavior is preserved. With the fix applied, the 100% pass rate of our quiche testing suite was restored. After the fix, cwnd grows along the expected CUBIC curve and the download completes in ~4-5 seconds. We don't worry about the losses at the end of the connection — that's expected because we fully utilized the router's allocated buffer. In other words, we are fully utilizing the available bandwidth in this test case. "Idle" is harder to define than it sounds. Normal pipeline delays at small windows can look like idleness to simple checks. Minimum-cwnd dynamics are a unique corner case. The bug was invisible at high speeds and only triggered after severe loss. The fix was surprisingly small compared to the complexity of the behavior. After weeks of instrumenting qlogs and analyzing visualizations to find the root cause, the solution required changing just three lines of code. As we noted during the investigation: the effort to find the bug was massive, but the fix itself was basically one line of logic. The fix described in this post has been contributed to cloudflare/quiche , Cloudflare's open-source implementation of QUIC and HTTP/3. Our CCA efforts go beyond loss-based algorithms: we also use quiche’s modular congestion control design to experiment with and tune our model-based BBRv3 implementation, now enabled for a growing percentage of our QUIC deployments. Stay tuned for further updates on QUIC congestion control implementation and performance. If you're interested in congestion control, transport protocols, or contributing to open-source networking code, check out the quiche repository. We're always looking for talented engineers who love digging into problems like these, please explore our https://www.cloudflare.com/careers/ open positions .